(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Biodiversity Heritage Library | Children's Library | Advanced Microdevices Manuals | Linear Circuits Manuals | Supertex Manuals | Sundry Manuals | Echelon Manuals | RCA Manuals | National Semiconductor Manuals | Hewlett Packard Manuals | Signetics Manuals | Fluke Manuals | Datel Manuals | Intersil Manuals | Zilog Manuals | Maxim Manuals | Dallas Semiconductor Manuals | Temperature Manuals | SGS Manuals | Quantum Electronics Manuals | STDBus Manuals | Texas Instruments Manuals | IBM Microsoft Manuals | Grammar Analysis | Harris Manuals | Arrow Manuals | Monolithic Memories Manuals | Intel Manuals | Fault Tolerance Manuals | Johns Hopkins University Commencement | PHOIBLE Online | International Rectifier Manuals | Rectifiers scrs Triacs Manuals | Standard Microsystems Manuals | Additional Collections | Control PID Fuzzy Logic Manuals | Densitron Manuals | Philips Manuals | The Andhra Pradesh Legislative Assembly Debates | Linear Technologies Manuals | Cermetek Manuals | Miscellaneous Manuals | Hitachi Manuals | The Video Box | Communication Manuals | Scenix Manuals | Motorola Manuals | Agilent Manuals
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "The floods of 1913 in the rivers of the Ohio and lower Mississippi valleys .."






GB 






■ 






■ 










V 



1^.4.* 






> 







■ 



I 









■ 



4) 4, ■ , 



■ 







(lass 9^/3 /*^ 

Book /fjtoj/ jTt 



W. B. No. 52B 



Issued December 81, 1913 



U. S. DEPARTMENT OF AGRICULTURE 

WEATHER BUREAU 
C. F. MARVIN, Chief 



THE FLOODS OF 1913 



IN THE 



RIVERS OF THE OHIO AND LOWER MISSISSIPPI VALLEYS 



Bulletin Z 



by 
ALFRED J. HENRY 

Professor of Meteorology 




WASHINGTON 

GOVERNMENT PRINTING OFFICE 

1913 



Issued December 31, 1913 

U. S. DEPARTMENT OF AGRICULTURE 

WEATHER BUREAU 
C. F. MARVIN, Chief 



THE FLOODS OF 1913 



IN THE 



RIVERS OF THE OHIO AND LOWER MISSISSIPPI VALLEYS 



Bulletin Z 



BY 



ALFRED j/ HENRY 

Professor of Meteorology 




WASHINGTON 

GOVERNMENT PRINTING OFFICE 

1913 



u I 



.PU US' 



o; of o, 

WAR 4 5914 







LETTEE OF TRANSMITTAL. 



United States Department of Agriculture, 

Weather Bureau, Office of the Chief, 

Washington, D. C, October 2, 1913. 
The honorable the Secretary of Agriculture. 

Sir : I have the honor to transmit herewith a report on the disastrous floods of March and 
April, 1913, in the States of Ohio and Indiana"; also of the resulting floods in the Ohio and 
lower Mississippi Eivers, together with tables and illustrations appertaining to the same. This 
report has been prepared by Prof. Alfred J. Henry. 

I recommend its publication as Bulletin Z of the Weather Bureau. 
Very respectfully, 

C. F. Marvin, Chief of Bureau. 
Approved. 

B. T. Galloway, Acting Secretary. 



TABLE OF CONTENTS. 

Page. 

Letter of transmittal 3 

Acknowledgments _ 9 

Floods in the Ohio River. 1870-1913 11 

The Basin _ H 

Contributing causes H 

During the last 40 years 12 

By groups 14 

Individual floods before period of systematic observations 14 

Floods of the later period . . . . 14 

Recapitulation of floods in the Ohio during the last 40 years. T5 

Ohio and Mississippi floods of January, 1913 ] 5 

The origin of the March, 1913, flood , _ \q 

Precipitation 16 

Excessive precipitation in Ohio 16 

The flood in the Ohio River, March, 1913 24 

In the headwaters. 24 

The discharge of the Ohio at Parkersburg, W. Va. 25 

From Cincinnati to Louisville . . 26 

From Louisville to Cairo 26 

Daily river gage readings in tributaries March 22-29, inclusive, 1913 27 

Previous floods in the Ohio River 29 

Comparison of 1884 and 1913 floods 30 

Precipitation, flood of 1884 31 

Are floods in the Ohio increasing 33 

Secular variation of precipitation ; _ 34 

Floods in the Ohio River by 15 and 20 year periods 36 

Conclusions as to flood frequency 36 

Probable maximum flood at Pittsburgh 37 

The flood in the lower Mississippi River, March-April, 1913 39 

Comparison with the flood of 1912 40 

Area overflowed 41 

Money loss due to floods of 1913 42 

Detailed reports on floods of 1913 , by districts 43 

Precipitation and floods in Ohio. March, 1913, by Prof. J. Warren Smith, Columbus, Ohio 44 

Precipitation, March, 1913. , 45 

Previous heavy rainfalls 46 

Rise of the rivers 46 

Loss and damage by the flood . 47 

The Maumee River. 47 

The Sandusky River 48 

The Mahoning River. 49 

The Hocking River 49 

The White Water River of Indiana 49 

The Great Miami River 50 

The flood at Dayton, Ohio, by H. C. Alps, Dayton, Ohio 51 

Precipitation 52 

Losses 53 

Dayton floods in past years ". 53 

Precipitation over Great Miami watershed. March 21-27, inclusive, 1913 54 

The flood below Dayton, Ohio. 54 

At Hamilton. Ohio. 55 



b CONTENTS. 

Detailed reports on floods of 1913, by districts — Continued. 

Precipitation and floods in Ohio, March, 1913 — Continued. p age . 

The Little Miami River 55 

The Scioto River 56 

The flood at Columbus, Ohio 57 

Below Columbus, Ohio 59 

The Muskingum River 59 

The flood at Zanesville, Ohio 60 

Below Zanesville, Ohio 61 

At McConnellsville, Ohio (by C. H. Morris, cooperative observer) <il 

At Marietta, Ohio 62 

Hourly rainfall in Ohio River watershed, March 23-26, 1913 63 

Mean daily rainfall and total for the period, March 23-27, 1913 64 

Daily river gage readings, March 24-28, 1913 65 

Special river gage readings, March 24—27, 1913 65 

Losses, tabulated by counties 67 

Flood in the "White River of Indiana. March, 1913, by C. E. Norquest, Indianapolis, Ind 71 

The White River Valley 71 

Heavy rainfall of March 23-27, 1913 71 

River stages 72 

Flood summary 73 

Loss of life and property 74 

Flood in the Wabash River of Indiana, March, 1913, by W. R. Cade, Terre Haute, Ind 75 

Flood in the Illinois River, by Montrose W. Hayes, St. Louis, Mo. 77 

The flood in the Ohio River in the Louisville district, by Prof. F. J. Walz, Louisville, Ky 81 

Precipitation 81 

Crest stages 81 

The Kentucky River 81 

The Ohio River 82 

Flood warnings 83 

The flood in the Ohio River in the Evansville district, by Al Brand, Evansville, Ind 85 

Its origin and progress 85 

Losses and damage 87 

The flood in the Mississippi River in the Memphis district, by S. C. Emery. Memphis, Tenn 89 

Its origin and progress 89 

Crevasses 90 

Flood warnings 92 

The floods in the Mississippi River in the Vicksburg district, by W. E. Barron, Vicksburg, Miss 93 

The January flood 93 

The March- April flood 93 

Flood warnings 94 

Losses and damage 96 

Floods in the Arkansas and White Rivers of Arkansas, by H. F. Alciatore, Little Rock, Ark 97 

Comparison of 1912 and 1913 floods 97 

Notes on the spring flood of 1913 on the Arkansas side of the Mississippi River 98 

Floods in the Mississippi River below Vicksburg, and in the Atchafalaya River in the spring of 1913, by 

I. M. Cline, New Orleans, La 101 

Warnings for the flood in the Mississippi River 102 

Action taken as a result of the warnings. 102 

Crevasses 102 

Area flooded 102 

Loss and damage 103 

The flood in the Hudson River, March, 1913, by George T. Todd, Albany. N. Y 105 

Floods in New York State (by R. E. Horton, Albany, N. Y.) 105 

Supplemental note on frequency of recurrence of Hudson River floods (by R. E. Horton, Albany, N. Y.). 109 

Floods in the Connecticut Valley and in Vermont, March, 1913, by W. W. Neifert, Hartford, Conn 113 



ILLUSTRATIONS. 



Charts. Page. 

1. Oliio River watershed, special rainfall chart, March 23-27, inclusive, 1913 28 

2. Ohio River watershed, total precipitation, January 1-13, inclusive, 1913 28 

3. Ohio River watershed, total precipitation, Otober 3-6, inclusive, 1910 28 

4. Ohio River watershed, total precipitation, February 4-7, inclusive, 1S84 2S 

5. Map of overflowed area in Mississippi Valley, 1913 41 

6. Total precipitation in Ohio, 8 a. in., March 23, to 8 a. in., March 25, inclusive, 1913 45 

7. Total precipitaton in Ohio for 48 hours ending 7 p. m., March 25, 1913 45 

8. Total precipitation in Ohio, March 23 to 27, inclusive, 1913 45 

9. Loss to bridges and highways in Ohio, by comities, March, 1913 45 

10. Overflowed area in Dayton, Ohio 52 

Diagrams. 

I. Comparison of crest stages in the Ohio in floods of 1884 and 1913 30 

II. Progressive averages of precipitation in the Ohio Valley 35 

III. Hydrograph, Great Miami at Dayton. Ohio, March 23-29. 1913 51 

IV. Hydrographs for floods of 1913 42 

Half tones — Figures. 

1. The Ohio in flood at Cincinnati, April 1, 1913, steamer Princess over river gage, foot of 

Broadway — Elevated railroad to right — Water three blocks back 25 

2. Wabash in flood at Mount Carniel, 111 26 

3. River front, Paducah, Ky., April, 1913 . 25 

4. Crevasse near Beulah, Miss., in February, 1913 40 

5. Trestle work and rock fill — Closing crevasse at Beulah, Miss., March, 1913 . 40 

6. Closing Beulah crevasse — Dumping dirt on river side of rock fill, March, 1913 40 

7. Closing Beulah crevasse — Land side of fill, 1913 40 

8. Closing Beulah crevasse — Both sides of fill. Note difference in height of water. 1913 40 

9. Fourth and main Streets, Dayton, Ohio — Flood of March, 1913. Looking north on Main Street — 

Water about 10 feet deep. Taken shortly before dark, Tuesday, March 25, 1913 (copyright 

by Smith Bros., Dayton. Ohio) 51 

10. Steele High School building, Dayton, Ohio — Showing destruction due to undermining foundation 

by water 51 

11. Washington Street Bridge (Indianapolis, Ind. ), looking downstream 73 

12. Indianapolis Water Co. — Weather Bureau river gage shelter, small round building left center 

of picture 73 

13. Threatened cave-in in main levee, Remy, La., and reenforcement of sand bags, April, 1913 102 

7 



ACKNOWLEDGMENTS. 



Special acknowledgments for valuable material furnished are due to District Forecaster 
I. M. Cline, New Orleans; Profs. J. Warren Smith, Columbus, Ohio, and F. J. Walz, Louis- 
ville. Ky. ; local forecasters, Penny witt, Pittsburgh; Devereaux, Cincinnati; Brand. Evansville; 
Lindley, Cairo; Hayes, St. Louis; Emery, Memphis; Barron, Vicksburg; Alps, Dayton; Todd, 
Albany; and Neifert, Hartford; section directors, Howe, Parkersburg; Alciatore, Little Rock; 
Church and Norquest, Indianapolis; and Cade, .Terre Haute. 

Also to Mr. R. E. Horton, of Albany, N. Y., for a report upon the floods in the State of 
New York. 

I am further indebted to Mr. W. O Devereaux, in charge of the Cincinnati Weather Bureau 
Office, for valuable suggestions and for reading the manuscript of Floods in the Ohio River; to 
Herman W. Smith and Andy M. Hamrick, of the river and flood division, for valuable 
assistance. 



Errata: Delete the words " Part 2 " on Charts Nos. 1 to 10. inclusive. 



FLOODS IN THE OHIO RIVER, 1870-1913. 



By Alfred J. Henry. 



The Ohio Basin. — The Ohio Basin is second in size of the six great natural divisions of the 
Mississippi Basin, yet it ranks first in importance in the causation of damaging floods in the 
larger stream. The topography of the basin in the western and northern portions is generally 
flat and rolling, but between those portions and the eastern and southern boundaries of the 
basin almost all conditions of surface contour may be found. It should be remembered that the 
eastern and southern boundaries, for the most part, lie along the crests of the Alleghenies and 
related mountain ranges, and that down the rugged western slopes of these mountains flow the 
streams which form the southern tributaries of such rivers as the Monongahela, the Little 
Kanawha, The Great Kanawha, the Big Sandy, the Kentucky, the Cumberland, and the Ten- 
nessee. On the headwaters of these rivers the slopes are steep, gradually becoming less as the 
lowlands are reached. 

Contributing causes of Ohio floods. — The Ohio is preeminently one of the turbulent rivers 
of the United States among streams of its size and drainage area. There are several important 
reasons why this should be so. First in order of importance is the accident of geographic loca- 
tion considered with respect to the meteorological conditions which dominate the weather of 
the interior of the continent. The longer dimension of the basin extends in a southwest- 
northeast direction from northeastern Mississippi to southwestern Xew York, a distance of 
about 800 miles; its shorter dimension stretches from northern Indiana southeastward to 
northern Georgia, a distance of about 500 miles: its total area is 201,700 square miles, and 
practically all of this vast area lies wholly within the region of frequent and copious rain- 
storms which, particularly in the winter and spring, pass from Texas to New England, directly 
over the longer axis of the basin of the river. Moreover, the northern portion of the basin. also 
lies within the area of rainfall produced by storms which pass across the continent from west 
to east over the Great Lakes. Owing to certain phases of storm development and movement 
not at present susceptible of satisfactory explanation, a storm passing eastward along or through 
the northern tier of States sometimes leaves an unsettled condition in its rear which may 
extend southwestward to the Gulf of Mexico, in which secondary storm centers are apt to 
develop and move northeastward through the lower Mississippi Valley, depositing heavy rains 
in both valleys. The northern portion of the Ohio Basin is, therefore, so located with respect 
to storm movement that it receives at times two downpours in quick succession. The torrential 
rains of March 23-27, 1913, illustrate this possibility, with the exception that the interval 
between the passage of the two storms on those dates was so short that the rainfall seemed to 
be, and was for all practical purposes, continuous. 

Another meteorological condition occasionally develops over the interior of the continent, 
technically known as the formation of a barometric trough, which separates two regions of 
higher pressure, one to the southeast and the other to the northwest : the southeastern or Atlantic 
area of high pressure extends well over the ocean, and the wind movement along and in its 
western borders is from the south and its moisture content is naturally high. The air of the 
northwestern or continental area of high pressure is cold and dry, and the opportunity is 
therefore continually present for the cold, dry air of the northwest to underrun and force 
upward the lighter moist air which belongs to the Atlantic high-pressure system. The quality 

11 



12 THE FLOODS OF 1913. 

or condition of immobility is largely developed in the latter, and to this quality may be attrib- 
uted the long-continued rains which frequently attend " trough " pressure formations over the 
Ohio and lower Mississippi Valleys. 

Primarily all floods are the results of falling rain or melting snow, or a combination of 
the two. The winter precipitation on the headwaters of the streams which take rise in the 
Alleghenies is mostly in the form of snow. Under favorable conditions as to temperature 
snow may accumulate in the mountains, and even on the lowlands, of the northern portion of 
the basin, until its presence becomes a menace to the dwellers in the lowlands, since there is 
ever present the possibility of a temporary warm spell or a warm rain occurring throughout 
the watershed. The transition from the lower warmer end of the basin to the upper colder 
portion is easity and quickly accomplished even in the heart of winter. 

As a natural result of the meterological conditions described in the preceding paragraphs, 
the Ohio Basin has a greater precipitation than any other of the northern tributary basins of 
the Mississippi. Finally, among other important contributory causes are the physical features 
of the basin in western Pennsylvania, "West Virginia, Kentucky, Tennessee, and western North 
Carolina and the extreme southwestern portion of Virginia, such as steep slope of stream beds, 
lack of surface storage, etc. The physical features are permanent and their influence upon 
stream flow is possible of determination. The character of the seasons, however, can not be 
foreseen; the necessity of being prepared for destructive floods during winter and spring is, 
therefore, ever present. 

Fortunately, all of the causes which contribute to the formation of floods in the Ohio Basin 
are seldom in operation over the entire watershed at one and the same time ; that is to say, all 
of those causes which, if operating unitedly, would form an unprecedented flood generally act 
disconnected^ 7 , and thus we experience floods less severe than might be expected under a more 
favorable conjunction of natural causes. Thus, as regards precipitation, it may be considered 
as a fundamental proposition that as the area of a watershed increases the probability of rain 
falling simultaneously over all portions of it diminishes. The full force of the proposition is 
better realized if we stop to consider the result of floods converging at Cairo, let us say, simul- 
taneously from the upper Mississippi, the Missouri, and the Ohio, and then to such a flood add 
the waters of the Arkansas, the St. Francis, and the Red, and we have a volume of water to 
combat staggering to contemplate. Such a combination of flood conditions has never been 
known to occur, but it is within the limits of reason to assume that atmospheric conditions 
favorable to a modified form of the above suggestion may occur. 

Ohio foods of the last 40 years. — In the last 40 years the Ohio River has been in severe 
flood seven times, using the expression " severe flood " to signify a stage of 5 feet or more above 
the flood stage for at least three-fifths of the distance between Pittsburgh and Cairo. 

The term " flood stage " is synonymous with another term, viz, " danger line," which was 
formerly used by the Weather Service and is still in current use in flood literature. The flood 
stage is a point at which the river overflows its banks and begins to damage property in 
proximity thereto. The flood stage at any point naturally depends on the height of the river 
banks above low water. At Pittsburgh it is 22 feet ; Parkersburg, 36 feet ; Cincinnati, 50 feet ; 
Louisville. 28 feet; Evansville, 35 feet; and Cairo, 45 feet. 

Technically, a river may be in flood when it reaches the adopted flood stage, but actually 
serious damage is in proportion to the height above the flood stage reached by the river. It is 
preferred to distinguish three classes of floods, viz, (1) technical floods or freshets, during which 
the river does not pass more than 1 foot above the flood stage; (2) severe floods, to signify 
stages from 2 to 5 feet above flood stage: and (3) great floods, to indicate the greatest recorded 
Hoods. If we take the maximum stage of water recorded at the several stations along the Ohio 
River as having a value of 1, we may express other floods in terms of the maximum stage; thus, 
5 feet above flood stage at points between Pittsburgh and Louisville is equivalent to a stage of 
0.7 to 0.8 of the maximum stage. Between Evansville and Cairo, 5 feet above the flood stage is 
equivalent to from 0.85 to 0.0 of the maximum stage. 



FLOODS IN THE OHIO RIVER, 1870-1913. 



13 



Severe floods in the Ohio River occurred in the years 1882, 1883, 1884, 1897, 1898, 1907, and 
1913: of these the floods of 1884, 1907. and 1913 may be classed as "great floods." Two severe 
floods occurred in each of the years 1907 and 1913. There were, of course, other years of flood 
on individual stretches of the river, and also other years when the river was in moderate flood 
from Pittsburgh to its junction with the Mississippi at Cairo. The subjoined table includes 
practically all of the severe floods, with the stages recorded at Pittsburgh. Parkersburg, Cin- 
cinnati, Louisville, Evansville, and Cairo. The criterion of a severe flood, as before stated, is 
a stage of 5 feet or more above flood stage, although a few stages a little short of 5 feet have 
been given for Cairo to complete the record of upriver floods. 



Table I. — Ohio River floods of 5 feet and more above flood stage. 1 



Year. 


Pittsburgh. 
(22 leet). 


Parkersburg 
(36 feet). 


Cincinnati 
(50 feet). 


Louisville 
(28 feet). 


Evansville 
(35 feet). 


Cairo 
(45 feet). 2 




Stage. 


Date. 


Stage. 


Date. 


Stage. 


Date. 


Stage. 


Date. 


Stage. 


Date. 


Stage. 


Date. 


1832 


35.0 


Feb. 10 


49.5 




64.2 
58.6 
66.3 
71.1 
56.3 
56.5 
59.2 
57.3 
54.9 
61.2 
61.4 
57.4 


Feb. 18 
Feb. 21 
Feb. 15 
Feb. 14 
Feb. 5 
Feb. 28 
Mar. 25 
Feb. 25 
Feb. 20 
Feb. 26 
Mar. 29 
Mar. 8 


40. S 
37.4 
44.4 
46.5 




46.3 
44.9 
47.8 
4S.0 
43.2 
43.9 
44.4 
42. S 
41.8 
43.6 
44.8 
42.7 
40.0 
42.4 








1882 


• 


Feb. 22 
Feb. 16 
Feb. 15 


Feb. 24 
Feb. 19 

...do 
Feb. 8 
Mar. 5 
Mar. 30 
Mar. 2 
Feb. 24 
Mar. 2 
Apr. 2 
Mar. 12 
Mar. 11 

...do 


51.9 
52.2 
51. S 
48.5 
48.8 
48.7 
46.2 
44.9 
51.6 
49.8 
46.2 


Feb. 26 


1883 


28.0 
33.3 


Feb. 8 
Feb. 6 


45.2 
53.9 


Feb. 10 
Feb. 9 


Do. 


1884 


Feb. 23 


1887 


Mar. 9 


1890 










34.1 
35. 5 


Mar. 3 
Mar. 2S 


Mar. 12 


1890 










Apr. 6 
Mar. 4 


1891 


31.3 


Feb. 18 


44.6 


Feb. 21 


1S93 






Feb. 28 


1897 


29.5 
2S.5 


Feb. 23 
Mar. 24 


37.9 

47.8 


Feb. 25 
Mar. 26 


35.4 
36.3 


Feb. 28 
Mar. 30 


Mar. 25 


1898 


Apr. 6 
Mar. 30 


1899 


1902 


32.4 
28. 9 
30.0 
29.0 


Mar. 1 
do 












1903 














50.6 


Mar. 15 


1904 


Jan. 23 

Mar. 22 


42.0 
42.4 
40.1 
51.6 
41.2 


Jan. 26 
Mar. 23 
Jan. 21 
Mar. 16 
Feb. IS 












1905 


















1907 


65.2 
62.1 


Jan. 21 
Mar. 18 


41.4 
36.0 


Jan. 22 
Mar. 20 


46.2 
43.8 
41.5 
43.2 
42.6 
46.7 
48.3 


Jan. 24 
Mar. 23 
Mar. 14 
Mar. 1 
Mar. 31 
Jan. 19 
Apr. 1 


50.4 
46.2 


Jan. 27 




35.5 
30.7 


Mar. 15 
Feb. 16 


Mar. 24 


1908 




1909 






33.0 


Feb. 27 


47.3 
54.0 
48.9 
54.8 


Mar 17 


1912 














Apr. 6 


1913 


31.3 
30.4 


Jan. 9 
Mar. 28 


45.1 
58.9 


Jan. 13 
Mar. 29 


62.2 
70.0 


Jan. 15 
Apr. 3 


39.5 
44.8 


Jan. 14 
Apr. ,2 


Jan. 26 


1913 


Apr. 4,7 





■ All of the stages in the above table except those of the 1832 flood are from observations bv the Weather Bureau. The 1832 stages have been col 
lected from various sources and are believed to be authentic. 

2 Some stages less than 5 feet above flood stage are given to complete the record of upriver floods. 

In the forty-odd years considered, the river at Pittsburgh was 5 feet or more above the flood 
stage on 13 separate occasions. Six of these floods continued throughout the course of the 
river to Cairo. The remaining seven, for one reason or another, dwindled away on the lower 
reaches of the river. Prominent among the causes of the decay of a flood is a lack of syn- 
chronism between the flood wave that is passing downstream and the output of the lower 
tributaries. There may have been sufficient precipitation over the watershed to cause a serious 
flood, but if the lower tributaries put out their quota of water either before or after the passage 
of the flood wave, the latter is much flattened. One of the most interesting cases of flood 
decay is that of the March (1907) flood, which gave the highest water ever recorded at Pittsburgh, 
Pa., a stage of 62.1 feet at Cincinnati, yet the river at Cairo barely passed the flood stage of 
45 feet, the final reading being 46.1 feet. The causes of this apparent anomaly are as follows : 
The excess water at Pittsburgh and along the upper reaches of the stream was due to a snow 
covering of from 4 to 8 inches which overlaid the watersheds of the Conemaugh, Kiskiminetas, 
and Youghiogheny Rivers. Rain and warm weather combined caused the greater portion 
of this snow to melt and run into the rivers and small streams, but the run-off from West 
Virginia rivers and other tributaries farther downstream depended entirely upon rainfall 



14 THE FLOODS OF 1913. 

and was naturally not extraordinary, since the snowfall which caused the great run-off in the 
Pittsburgh district was not widespread. As a consequence the volume of water entering the 
river from its southern tributaries was not sufficient to maintain the stream at record-breaking 
stages. Another factor which must be taken into consideration when estimating the Cairo 
stage which should result from an upstream flood wave is the stage of the Mississippi River 
at Cairo at the time the flood waters of the Ohio discharge into it. since a full Mississippi 
has the effect of backing up the Ohio and causing a higher reading on the Cairo gage than 
the same volume of water would with the Mississippi at a low stage. In the March (1907) 
flood in the Ohio, the Mississippi was considerably lower than during the January flood of 
the same year. The details of the above-named flood illustrate the necessity of making a 
separate study of each flood, and particularly as to the circumstances of its origin and de- 
velopment. 

Ohio -floods by groups. — For convenience of discussion we may assign severe Ohio floods 
to one of the three following groups according to their origin : First group; floods in midwinter 
and spring caused by extensive and continued rains during an open winter. The flood of 
January, 1913, is typical of this group. Second group; floods caused by a short period of 
rain coming in conjunction with a midwinter thaw. The February (1884) flood is typical of 
the second group. We are compelled to make a third group, of which we have the single 
example of March, 1913. The third group therefore consists of floods caused by torrential 
rains extending over a comparatively short period, 72 to 96 hours. Other floods partake of 
some of the characteristics of each group and are therefore difficult of classification. The 
melting of the winter's snow is generally a factor, but its influence sometimes is practically 
negligible. 

Individual Ohio foods before the period of systematic observations. — We now mention 
very briefly Ohio floods within historic periods. 

Flood of 1806: The Ohio at Pittsburgh reached a stage of 33.9 feet. Independent con- 
firmation of this flood may be found in " History of Clarion Co., 1887,'' where mention is 
made of a great flood in Red Bank Creek in 1806. Red Bank is one of the eastern tributaries 
of the Allegheny River, which it enters south of the Clarion River. 

Flood of 1832 : This flood furnished a stage on the Pittsburgh gage of 35 feet, which endured 
as the highest of record for 75 years ; the flood was caused by rain falling upon frozen ground, 
melting what snow there was and running off as fast as it fell. The 1832 flood was of con- 
siderable magnitude at Cincinnati and ranks as one of the great floods at that place. Maxi- 
mum stages reached by this flood at Louisville. Ivy., and Evansville, Ind., are given in Table I, 
and the following comparative statement of flood crests in early times is due to Mr. Adam 
Crozier, one of the early cooperating observers of the Smithsonian Institution and later of the 
Signal Service. 

Speaking of the flood of 1882, he says, under date of February 22, 1882: 

The Ohio River reached its highest point here (35 miles below Louisville) to-day. Compared with that 
of former great floods it was as follows: Thirty-four inches below the flood of 1847; 33 inches above the flood 
of 1S53 ; and 1 inch above the flood of 1867. 

Inasmuch as the 1847 flood is said to have been 2 feet under that of 1832, the last-named 
flood must also be counted among the first magnitude floods in the lower stretches of the river, 
although it has been overtopped by the floods of 1883, 1884, 1907, and 1913. 

Flood of 1847 : This was a severe flood at Cincinnati and was probably due to floods in the 
West Virginia and Kentucky tributaries, since the stages both above and below Cincinnati indi- 
cate a moderate flood only. 

There were also floods in the early fifties and the early sixties, but little definite informa- 
tion of them has been preserved beyond a record of the maximum stages at Pittsburgh and a 
few other upriver points. 

Floods of the later period. — The records of the Army Engineers and the Signal Service, 
now Weather Bureau, include daily river stages at principal points along the Ohio from 1870 



FLOODS IN THE OHIO RIVER, 1870-1913. 15 

to date. At Pittsburgh and Cincinnati daily gage readings are available for an earlier 
period; otherwise the record begins in 1870. In the decade 1870-1880 there were no general 
floods of consequence, but, coincident with an increase in precipitation in the early eighties, a 
series of floods set in beginning in 1882, continuing through 1883, and culminating in the great 
Ohio flood of 1884. In the same decade there were also minor floods in the lower part of 
the river in 1887. 

In the decade 1890-1900 there were general floods in the years 1S97 and 1898, but they were 
not unusually severe. 

In the 10 years 1900 to 1910 there were numerous minor Hoods in the upper stretches of the 
river and also two severe floods in 1907. 

A severe flood in the lower stretches of the river in 1912 was followed by a general flood in 
January. 1913, and a greater flood in March-April of the same year. 

To recapitulate : The Ohio Eiver during the last forty-odd years has been in decided flood 
in its entirety seven times; it has been in discontinuous flood much more frequently; thus at 
Pittsburgh in the same period the number of times with a stage of 5 feet above the adopted 
flood stage of 22 feet was 13; owing to the fact that some of the most important tributary 
streams in flood causation enter the Ohio between Pittsburgh and Cincinnati, it would probably 
be a better criterion of great floods to use the stages recorded at Cincinnati instead of Pitts- 
burgh. Table I shows that a stage of the river at that place 5 or more feet above the flood 
stage, 50 feet, has occurred fifteen times; of these, however, four, viz, those of 1891, 1893, 1899, 
and the second flood of 1907, did not reach Cairo with an excess of at least 3 feet above flood stage, 
and must be classed as discontinuous or decaying floods; 11 of the 15 floods, however, con- 
tinued to Cairo with little or no abatement, and that number is believed to represent the flood 
frequency during the period 1871-1913. The years in sequence are 1882, 1883, 1884, 1890 (2), 
1897, 1898, 1907 (January), 1913 (2), and the classification according to that herein proposed 
is as follows: 

Group 1: Floods due to accumulated rains in January, February, and March 7 

Group 2: Floods due to winter rains in conjunction with a thaw 3 

Croup 3: Heavy rains within a few days 1 

Total Ohio floods, Cincinnati to Cairo 11 

The great majority of floods, therefore (7 out of 11), were due to heavy rainfall in the 
Ohio Basin ; severe floods occur almost invariably in the months of January, February, and 
March, sometimes lapping over into April. 

Passing now to a consideration of the floods of 1913, it may be remarked at the outset that 
the atmospheric conditions during January were unusually conducive to flood formation. A 
flood of moderate severity passed down the Ohio River beginning at Pittsburgh, Pa., on Jan- 
uary 9, reaching Cairo 17 clays later, and New Orleans, La., by February 23 — 45 days from 
Pittsburgh. The small table shows the progress of the January flood from Pittsburgh to New 
Orleans. 

Ohio and Mississippi River flood of January-February, ]!)13. 



Stations. 



Pittsburgh.. 
Parkersburg. 
Cincinnati... 
Louisville... 
Evans ville.. 
Cairo 



Flood 
stage. 



Feet. 



Crest 



Feet. 
31.3 
45.1 
62.2 
39.5 
46.7 
48.9 



Date 

Januarv, 
1913. 



Number 

of days 

above 

flood 



Stations. 



Flood 
stage. 



Mempliis 

Vieksburg 

Natchez 

Baton Rouge. . 
Donaldsonville 
New Orleans. .. 



Feet. 



Crest 
stage. 


Date 
Feb- 
ruary, 
1913. 


Feet. 




40.5 


3 


49.0 


16-18 


48.7 


20 


37.2 


22 


29.3 


22 


18.4 


23 



Number 
of days 

above 

flood 



25 
22 
21 
19 
17 
10 



16 



THE FLOODS OF 1913. 



The origin of the March, 1913, food. — The meteorological conditions which led up to the 
disastrous floods of March, 1913, in Ohio and Indiana, have been reported upon elsewhere by 
the writer (Monthly Weather Review, March, 1913) ; suffice it for our present purpose to 
summarize them very briefly in the following paragraph: 

An unprecedented amount of rain fell over the States of Ohio and Indiana in the space of 
72 hours, or from the afternoon of March 23 to the afternoon of March 26, 1913. The rain 
continued over these and adjoining States with practically no intermission, although at a lower 
rate of intensity during the 27th. the aggregate amount for the four-day period being absolutely 
without precedent for a like period over a considerable portion of the watershed of the northern 
tributary streams in Ohio and Indiana. The extension of the region of rainfall over the southern 
tributaries of the Ohio on March 26 and 27 made a flood in the streams of that region certain. 

EXCESSIVE PRECIPITATION IN OHIO. 

In order to maice a comparison of the accumulated amounts of precipitation in the case of 
the March, 1913, floods with previous storms, we have examined the continuous record of daily 
precipitation for Cincinnati, Ohio, for the period 1871-1913. The criterion on which the table 
below was formed was the occurrence of at least 1 inch of precipitation in 21 hours: that amount 
must have occurred at least in 1 of the 24 hours considered, hence each group of dates repre- 
sents the occurrence of at least 1 inch in 24 hours. If no rain fell on the day immediately 
preceding or subsequent to the date of heavy rain, the entry in the table is confined to a single 
date; if two or more dates are given, the second entry of rainfall represents the accumulated 
amount for 48 hours, the third the accumulated rainfall for 72 hours, etc. The arrangement 
of the table is chronologically by months. This arrangement permits us to note that the seasonal 
variation of excessive rains is not especially well marked; the principal maximum occurs in the 
months of May, June, July, and August, while there is also a prominent midwinter maximum 
in February. The minimum falls in September and October, as might be expected. 

Accumulated amounts of excessive precipitation at Cincinnati, Ohio, 1871 to June 30. 1913. 



Year. 



Day. 



24 
hours. 



48 
hours. 



72 
hours. 



96 

hours. 



Year. 



Day. 



24 
hours. 



48 
hours. 



72 
hours. 



96 
hours. 



JANUARY. 



1874. 
1876. 
1876. 
1876. 
1877. 
1880. 
1881. 



1885 . 
1890. 
1890. 
1891. 



871. 
S73. 
874. 
874. 
881. 
882. 
882. 
883. 
883. 



884. 



6- 7 

18 

22-23 

27-28 


0.72 

2.97 

.70 

1.18 


1.92 










1.90 
3. 53 










15-16 


1.09 


1.12 






8 

31 

4- 7 


1.16 
1.27 
1.05 














1.05 


1.94 


2.52 


15 


1.46 


2.65 


2.73 




4- 7 


.28 


.34 


1.64 


2.44 


15 


1.32 








31 


1.31 















1893. 
1895. 
1897. 



1898.. 
1898.. 
1899.. 
1906.. 
19071. 
1910.. 
1913 . . 
1913.. 



FEBRUARY. 



17-18 
15-16 
6- 7 
20-23 


1.50 
.31 

1.04 
.43 


2.00 
1.84 
1.11 
3.16 














3.79 


3.82 


7-10 


1.00 


2.01 


2.92 


3.01 


12-13 


.18 


1.48 






19-21 


1.00 


2.85 


3. 31 




4- 7 


1.22 


1.24 


2.75 


3.20 


10-11 
23-24 

4- 7 


1.65 
1.01 
1.35 


1.96 
1.10 
2.91 










4.56 


4.79 


10-13 


.14 


.73 


.79 


1.97 


19 


1.12 















1885.. 
1887.. 
1887.. 
1893 . . 
1894 . . 
1897.. 
1903 . . 
1904 . . 
1908.. 
1908.. 
1909.. 
1910.. 



1 

5- 7 

J- 5 

9-10 

19-20 

22-23 

13-14 

1- 2 

1- 3 

12-14 

5- 8 

10-12 



1.35 
.11 

.47 

1.57 

.68 

1.32 

.93 

.11 

1.08 

.19 

.14 

.70 



1.37 
1.64 
1.56 
1.84 
2.35 
1.55 
2.00 
1.40 
2.63 
2.66 
.70 
2.48 



3.92 



2.71 
2.68 
2.22 
2.53 



2.26 



8- 9 

2- 3 

26 

14-15 

12 

20-22 

15-16 

6- 7 

4- 5 

13-15 

23-24 

26-27 



0.24 

1.97 

1.19 

1.05 

1.14 

.45 

1.09 

.03 

.10 

.06 

2.69 

.07 



1.42 
3.31 



1.76 
1.79 
1.08 
1.41 
1.90 
2.70 
1.97 



3.21 



1.92 



1.94 



1 Continuous rain, 11-20; total, 4.37 in. 



EXCESSIVE PRECIPITATION IN OHIO. 17 

Accumulated amounts of excessive precipitation at Cincinnati, Ohio, 1871 to June 30. 191S — Continued. 



Year. 



Day. 


24 48 
hours, hours. 


72 
hours. 


96 
hours. 



Year. 



Day. 


24 

hours. 


48 
hours. 


72 
hours. 



96 
hours. 



MARCH. 



1871 


2- 3 

4- 7 

1- 3 

28 

12-14 

22 

20-21 

28-30 

20-21 

26-27 

30 


1.00 

.40 

1.02 

1.04 

1.82 

1.70 

2.54 

.45 

.13 

1.21 

1.08 


1.40 

.41 

1.05 






1897 


2- 5 

16-19 

28 

6- 8 
21-22 
25-26 
19 
29-30 
12-13 

8- 9 
24-27 


0.44 

1.07 

1.07 

.03 

.23 

.75 

.66 
2.00 

.63 
2.21 


0.75 
1.10 


0.75 
1.78 


5.72 




]s7-l 


1.88 
1.31 


1.93 


1897 


1.82 


1S75 


1899 

1903 




1876 


1.39 
2.25 

2.82 

1.90 
6.59 
1.88 
6.36 


2.49 




1878 


1.84 


1.93 




1904 

1904 

1906 

1906 

1907 

1909 

1913 




1879 






1882 


2.85 

.95 

1.36 

1.50 










1883 


2.45 






1888 






1891 










1896 






7.47 















APRIL. 



1872 


8- 9 
15 
13-15 
15-16 
24-26 
22-23 
17-18 
21-23 
24-27 
18-21 


1.62 

1.10 

1.90 

.87 

.36 

1.02 

.43 

.06 

.28 

1.72 


1.78 






1893 


9-11 
28-29 

7- 8 
25-27 
20-21 
23-26 
11-14 
26 

8-10 


0.23 

.41 

.08 
1.19 

.23 
1.15 

.31 
1.58 

.22 


1.04 
1.69 
1.10 
1.38 
1.25 
1.15 
.42 


2.38 


2.41 


1872 . 






1902 




1876 . 


1.99 
2.33 
1.80 
1.13 
2.36 
1.82 
.33 
1.73 


2.20 




1903 






1880 


1904 


1.52 




1880 


2.36 




1905 




1884 


1907 


2.20 
1.60 


2.51 


1887 . 






1911 


1.98 


1887 


2.80 
1.42 
2.90 


1.49 
3.18 


1912 




1890 


1913 


1.23 


2.14 




1892 









MAY. 



1871 


30 
17-18 

9-10 
25-27 
29-31 

9-11 
14 
27-28 
19-22 
26-29 
4 
11-13 
30-31 

7- 9 
29-31 


2.00 

1.03 

1.41 
.53 

1.76 
.2S 

1.21 
.46 
.33 
.18 

1.43 
.92 

1.00 
.11 
.42 








1893 

1893 

1893 

1899 

1900 

1902 

1903 

1903 

1904 


1 

9 

25-26 

' 29-31 

9-10 

21-24 

21-22 

27-30 

30-31 

11-14 

3- 6 

30-31 

5- 8 

28-29 


2.43 
1.12 
2.37 
1.12 

.40 
2.37 

.08 
1.13 
1.32 
3.16 

.68 
1.14 
1.24 

.23 








1S72 '. 


2.39 
1.57 
3.31 
2.23 
.40 












1875 






2.38 
1.15 
2.07 
2.37 
1.15 
1.42 
2.40 
3.35 
1.74 
1.17 
1.65 
1.43 






1S79 


3.34 

2.25 
2.42 




1.84 




1880 




1880 

1881 


2.37 


4.82 


1882 


2.62 
.71 
.53 






1.58 


1.96 


1883 


2.47 
.94 


2.52 
2.06 




1883 


1905 


3.75 
2.43 


5.06 


1884 


1908 

1911 

1912 

1912 


3.91 


1886 


2.16 
1.03 
1.17 
1.54 


2.67 






1887 


1.92 


2.01 


1888 


1.25 
1.62 






1889 











JUNE. 



1872 


12-14 

21-22 

1- 3 

9 

23 

24-26 

7-10 

8-11 

27-29 

9 

13-15 

25-26 

28-29 

8 


0.05 
.13 
.30 

1.28 

1.51 
.06 
.22 
.09 

2.05 

1.42 
.90 

1.96 
.50 

1.35 


1.15 
1.13 
1.59 


1.63 




1881 ." 


29 
9 

5- 7 
10-13 
14-17 
21-22 
23-24 
24-25 
27-30 

4- 6 
25-26 
21-24 

8-10 


1.00 
1.68 
1.10 
1.18 
.10 
1.00 
1.64 
1.22 
.05 
.03 
.34 
.26 
.91 








1875 


1886 

1887.. 








1876 


1.70 




1.79 
1.76 
1.65 
2.05 
1.69 
1.39 
1.80 
1.75 
2.52 
1.30 
1.05 


1.82 
2.26 
2.43 




1876 


1890 

1890 

1893 


2 85 


1876 








2.51 


1877 


1.44 

.69 

1.55 

2.61 


1.46 
2.70 
1.61 
2.63 


2.83 
1.71 




1878 


1896 . . 






1879 


1899.. 






1879 


1902 

1903 


2.64 
1.88 


2 75 


1880 


2.06 


1880 


3.52 
2.55 
1.54 


4.04 




1906.. 




\ 
1880 


1907 

1909 . 


1.30 
2.22 


1.83 


1880 








1881 





















14284°— 13- 



18 THE FLOODS OF 1913. 

Accumulated amounts of excessive precipitation at Cincinnati, Ohio, 1S71 to June SO, 1913 — -Continued. 



Year. 


Day. 


24 
hours. 


48 

hours. 


72 
hours. 


96 

hours. 


Year. 


Day. 


24 
hours. 


48 
hours. 


72 
hours. 


96 
hours. 










JULY. 















1872. 
1874. 
1S75. 
1875. 
1877. 
1877. 
1878. 
1880. 
1881. 
1889. 
1891. 
1891. 
1893. 
1896. 
1896. 

1871. 
1871. 
1871. 
1873. 
1873. 
1875. 
1876. 
1876. 
1878. 
1879. 
1879. 
1880. 
1882. 
1882. 
1885. 
1885. 
1886. 



15-17 

10-11 

22-23 

27-30 

18 

26-27 

13 

19 

14 

19 

7- 8 

30 

26 

9 

20-21 


0.14 

.17 
1.17 

.43 
1.24 

.24 
1.55 
1.54 
1.54 
2.40 
1.46 
1.59 
1.90 
1.18 

.06 


1.62 
1.87 
1.27 
1.63 


2.57 








2.30 


2.61 


1.30 




























2.43 






















2.08 











1896. 
1897. 
1897. 
1897. 
1898. 
1900. 
1901. 
1902. 
1903. 
1906. 
1907. 
1910. 
1911. 
1912. 



23-24 

5- 6 
24 
26 
31 
25 

30-31 

18 

22 

22-23 

9-11 

16-17 

6- 8 
17-18 



0.07 
2.20 
2.39 
1.01 
1.00 
1.04 
1.16 
1.39 
1.18 
2.16 
.16 
1.07 
1.43 
2.52 



1.08 
2.24 



3.45 
.99 
1.98 
2.61 
2.54 



3.59 



AUGUST. 



7- 8 
11 


1.40 
2.00 


1.45 










24-27 


.10 


2.40 


2.55 


2.60 


11-12 


.05 


1.40 






25-26 
1- 3 


.12 
2.03 


1.41 
2.21 






2.22 




5- 8 


1.89 


2.09 


2.34 


2.73 


16-18 


.10 


1.75 


2.32 




18-21 


1.22 


1.47 


2.07 


2.09 


4- 7 


.06 


1.93 


2.83 


3.94 


23-25 


1.04 


2.14 


3.81 




1- 3 


.27 


.60 


2.17 




23-25 


1.17 


1.37 


1.38 




26-29 


.72 


2.60 


2.79 


2.82 


6- 7 
22-25 


.90 
1.05 


2.62 
1.06 






1.07 


1.44 


15-17 


.55 


1.72 


1.73 





1890. 
1890. 
1892. 
1894. 
1895. 
1896. 
1898. 
1899. 
1899. 
1903. 
1905. 
1906. 
1906. 
1912. 
1912. 



20-21 
18-21 
25-27 
11 
10-13 
26-27 

1- 2 
18-19 

5- 6 
10-12 
25-27 
19 
17-18 
25-27 

8-11 
19-20 



0.26 

1.12 

.05 

1.06 

.14 

.81 

1.89 

1.18 

1.51 

.09 

.14 

1.37 

.04 

.08 

.09 

.12 



2.72 
1.61 
2.60 



1.34 
1.98, 

1.90 
1.21 
1.98 
1.49 
1.25 



1.34 
1.53 
2.02 

1.51 



1.68 
2.71 



1.52 



1.55 
1.36 



1.71 
2.19 



2.06 



1.38 



SEPTEMBER. 



1873 


22 
5 

1- 3 
12-13 
23-24 

8 

2- 5 

3- 6 
13-14 
11-14 

30 


1.12 
1.14 

.06 
1.13 

.35 
1.25 
1.50 
1.28 
2.02 

.15 
1.60 








1894 


15-18 

27-30 

22 

7 
6- 7 
1 
9-11 
26-29 
5 
8-11 
14-16 


0.12 
.68 

1.24 

1.22 
.16 

1.04 
.08 
.72 

2.16 
.04 
.22 


1.32 
1.93 


1.75 
3.67 


1.82 


1874 








1896 


3.98 


1879 


2.05 
1.28 
1.66 


2.57 




1898 




1879 


1899 








1884 






1900 


1.24 






1885 






1902 . 






1889 


1.71 

1.68 

2.10 

.48 


1.90 
2.13 


2.14 
2.14 


1903. . 


1.13 
.81 


1.24 

.83 




1891 


1906... 


1.96 


1892 


1911 




1893 


1.58 


2.19 


1911 


.15 
1.23 


1.49 
1.63 


3.02 


1893 


1911 . . 

















OCTOBER. 



1876 


22-24 

2- 4 

13 

1- 2 

28-29 


0.76 
.36 

1.01 
.16 

2.15 


2.87 
2.29 


2.94 
2.43 




1893 . 


3- 4 
6- 7 

3 
10-11 

4- 6 


1.42 

.16 

1.93 

1.24 

.26 


1.44 
1.24 






1881 


1900 






1882 


1907... 






1883 


1.72 
3.99 






1909 


1.26 
2.59 






1883 






1910 . 


5.41 















EXCESSIVE PRECIPITATION IN OHIO. 19 

Accumulated amounts of excessive precipitation at Cincinnati, Ohio, is7i to June SO, 1918 — Continued. 



Year. 



Day. 



2-1 

hours. 



48 
hours. 



72 
hours. 



hours. 



Year. 



IViN 



24 

hours. 



•is 
hours. 



72 
hours. 



hours. 



NOVEMBER. 



1871. 
1S73. 
1874. 
1874. 
1875. 
1878. 
1879. 
1880. 
1881. 
1883. 
1883. 
1886. 



13-14 

21-24 

16-17 

22-23 

13-14 

27 

14 

4- 6 

17-19 

10 

21-22 

22-23 

8- 9 


1.40 
1.34 

.21 
1.82 
1.79 
1.35 
1.06 

.72 

.15 
1.04 
1.81 

.22 
1.0S 


2.21 
1.69 
1.29 
1.90 
2.29 
































.73 

1.37 


2.23 
2.18 




2.94 
1.42 
2.20 



















1889. 
1891. 
1891. 
1895. 
1896. 
1897. 
1897. 
1S98. 
1899. 
1900. 
1900. 
1910. 
1911. 



10 
22-24 

8- 9 
27-28 

1- 2 
25-26 
10-11 
21-23 
20-23 
24-26 
27 

6- 7 



0. n-s 

1. 10 
1.39 

.56 
1.38 
1 . 65 

.12 
1.34 

.06 
1.52 
1.45 
1.06 

.07 



1.10 



2.25 
1.98 
1.52 
2.73 
1.27 
1.92 
1.08 
1.80 
1.90 



1.08 



1.21 



2.29 



1.09 
2.07 
2.35 



2.13 
2.55 



DECEMBER. 



1870... 


19 

31 

3- 4 

27-29 

24-26 

9-10 

22-24 

1- 5 

14 

20-23 

5- 6 


1.00 
2.50 
2.47 

.06 
1.68 

.16 
1.24 
2.36 
1.12 
1.18 
1.0S 








1883 


22-24 
11-12 
3 
25-26 
11-12 

8-10 

14 

15-16 

24-27 

1- 2 
27-30 


0.69 
1.16 
1.35 


3.26 
1.43 


3.71 




1871.. 








1884 




1873 


2.53 
1.13 
1.71 
1.43 
2.42 
3.17 






1893 






1874 


1.27 
2.62 




1895 


.13 

1.16 
.03 
1.52 
1.31 
.43 
.15 
.12 


1.52 
1.36 
1.23 






1875 


1899 






1879 


1901 


1.30 




1879 


3.65 




1901 




1880 


1902 


1.53 
.77 

1.38 
.19 






1881 






1904 


1.97 


3.00 


1881 


2.09 
1.14 


2.86 


2.94 


1905 




1882 


1910 


1.28 


1.29 













The statistics of the above table are summarized in the small table. below, wherein is shown 
the maximum rainfall for each month, classified according to the length of the period, whether 
24, 48, 72, or 96 hours. From this table we see that the greatest 24-hourly amount in the forty- 
odd years, 1871-1913, was 3.32 inches in May, 1905 ; the greatest 48-hourty amount was 6.6 inches 
in March, 1907; the greatest 72-hourly amount was 7.5 inches in March, 1913; and the greatest 
96-hourly amount was 5.7 inches in March, 1897. 

Accumulated precipitation in 2%, 48, 72, and 96 hours at Cincinnati, Ohio {maximum for each month and for 

the year). 

|In inches.] 





Jan. 


Feb. 


Mar. 


Apr. 


May. 


June. 


July. 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


Max. 




3.0 
3.5 
3.9 
2.5 


2.7 
3.3 
4.6 
4.8 


2.5 
6.6 

7.5 
5.7 


1.9 
2.4 
2.9 
3.2 


3.2 
3.3 
3.7 
5.0 


2.0 
3.5 
4.0 
2.8 


2.5 
3.4 
3.6 
2.6 


2.0 
2.7 
3.8 
3.9 


2.2 
2.1 
2.6 
4.0 


2.2 
4.0 
5.4 


1.8 
2.9 
2.3 
2.5 


2.5 
3.3 
3.6 
3.0 


3.2 




6.6 




7.5 




5.7 








24 


25 


22 


19 


29 


27 


29 


33 


22 


10 


26 


22 


288 







The record of 7.5 inches in 72 hours in the March, 1913, rainfall is easily above any other 
record for a like period, the nearest approach to it being 5.4 inches in October, 1910. The figures 
in the table also express the principle elsewhere referred to in this paper, viz, that the intensity 
of precipitation varies inversely as the duration. In 7 out of 12 months the accumulated 
amounts for 96 hours are less than for a shorter time. 

The chronological variation of excessive rains, as determined from the number of excessive 
rains in each year (not published), seems to follow rather closely the variation in the yearly 



20 



THE FLOODS OF 1913. 



amounts of precipitation; thus, there was a maximum number of occurrences of excessive rain- 
falls in the years 1879-83, inclusive, which period was also one of abundant rains; conversely, 
years of small rainfall do not }'ield many occurrences of excessive precipitation, but there does 
not seem to be any relation between the amount of the excess of the j^early precipitation above 
normal and the number of occurrences of excessive rains. 

The details of the floods in Ohio and Indiana and the course of the wave of water which 
later passed down river to the Gulf of Mexico are now matters of history and will not be 
repeated here; it is our purpose, however, to present the statistics of this remarkable flood 
rather fully and to draw some comparisons with previous floods. Beginning with the precipi- 
tation, we present the numerical values of the daily amounts over the Ohio Basin in Table 
Xo. 2, below. 

In Table No. 2 the stations in Ohio and Kentucky are located in the various watersheds 
as follows : 

Ohio. — Antwerp, Benton Eidge, Toledo, and Wauseon in the Maumee watershed; Fre- 
mont, Tiffin, and Upper Sandusky in the Sandusky watershed; Bowling Green, Cleveland. 
Conneaut, and Hudson in the Lake Erie watershed; Cincinnati, Millport, and Portsmouth 
along the Ohio; Garrettsville in the Mahoning watershed; Belief ontaine, Dayton, Greenville, 
and Urbana in the Great Miami watershed : Kings Mills in the Little Miami watershed ; Circle- 
vine, Columbus, Frankfort, and Marion in the Scioto watershed; and Bangorville, Cambridge, 
Canal Dover, Canton, Granville. Philo, and Wooster in the Muskingum watershed. 

Kentucky. — Falmouth and Scott in the Licking watershed; Pikeville and Catlettsburg in 
the Big Sandy watershed; Berea, Beattyville, Shelby City, High Bridge, Lexington, and 
Frankfort in the Kentucky watershed ; Edmonton. Franklin, and Calhoun in the Green water- 
shed; Middlesboro, Williamsburg, Burnside, and Hopkinsville in the Cumberland watershed: 
and Anchorage, Maysville, Louisville, and Marion along the Ohio. 

Table Xo. 2. — Daily amounts of precipitation (in inches and hundredths) at representative stations in the 

watershed of the Ohio River, Mar. 23-27, 1913} 



"WESTERN PENNSYLVANIA. 



Regular stations (midnight to midnight): 

Pittsburgh 

Erie 

Special river stations (7 a. m. to 7 a. nO: 

Beaver Falls 

Confluence 

Ellwood City 

Franklin 

Freeport 

Greensboro 

Lock No. 4 

Mosgrove 

Parker 

Saltsburg 

Sharon 

Warren 

West Newton 

Cooperative stations (~ p. m. to 7 p. m.): 

Aleppo 

Claysville 

Clearfield 

Greenville 

Indiana 

Saegerstown 



Mar 23. 



0.20 
1.28 

0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.01 
0.00 
0.00 
0.00 
0.00 

0.10 

0.15 
0.00 
1.34 
0.23 
1.00 



Mar. 24. 



0.72 
1.38 

0.59 
0.00 
0.69 
1.41 
0.14 
0.06 
0.08 
0.19 
0.62 
0.05 
1.19 
0.70 
0.10 

0.08 
0.17 
0.64 
1.11 
0.62 
1.09 



Mar. 25. 



0.55 
2.12 

1.65 
0.00 
1.81 
2.22 
1.53 
0.04 
0.05 
1.77 
2.25 
0.20 
2.92 
1.70 
0.07 

0.58 
0.58 
0.31 
3.74 
0.26 
3.70 



Mar. 26. 



1.63 
0.91 

1.79 
1.00 
1.61 
1.32 
1.15 
1.40 
1.50 
1.93 
1.40 
1.50 
1.24 
1.36 
1.62 

1.09 
1.51 
1.09 
0.95 
1.17 
0.74 



Mar. 27. 



0.38 
0.58 

0.92 
0.76 
0.92 
0.88 
0.35 
1.12 
0.64 
0.67 
1.25 
0.77 
0.84 
1.10 
0.64 

0.97 
1.02 
0.90 
0.60 
0.80 
0.98 



Total. 



3.51 
6.27 

4.95 
1.76 
5.03 
5.83 
3.17 
2.62 
2.27 
4.56 
5.53 
2.52 
6.19 
4.86 
2.43 

2.82 
3.43 
2.94 
7.74 
3.08 
7.51 



1 Important.— The daily amounts of precipitation at regular Weather Bureau stations are given from midnight to midnight, seventy-fifth meridian 
time. At special river stations the amounts include the precipitation which occurred between 7 a. m. of one day and 7 a. m. of the following day, 
local time, while the amounts at cooperative stations include the precipitation from 7 p. m. of one day to 7 p. m. of the following day, local time. 



PRECIPITATION IN OHIO BIVER WATERSHED MARCH 23-27, 1913. 



21 



Table Xo. 2. — Daily amounts of precipitation (in inches and hundredths) at representative stations in the 
watershed of the Ohio River, Mar. 23-27, 1913— Continued. 



Mar. 23. 



Mar. 24. 



Mar. 25. 



Mar. 26. 



Mar. 27. 



Total. 



OHIO. 

Regular stations (midnight to midnight): 

Cincinnati 

Cleveland 

Columbus 

Toledo 

Special river stations (7 a. m. to 7 a. m.): 

Kings Mills 

Portsmouth 

Upper Sandusky 

Cooperative stations (7 p. m. to 7 p. m.): 

Antwerp 

Bangorville 

Bellefontaine 

Benton Ridge 

Bowling Green 

Cambridge 

Canal Dover 

Canton 

Circleville 

Conneaut 

Dayton 

Frankfort 

Fremont 

Garrettsville 

Granville 

Greenville 

Hudson 

Marion 

Millport 

Philo 

Tiffin 

Tj rbana 

Wauseon 

Wooster 

WEST VIRGINIA 

Regular stations (midnight to midnight): 

Elkins 

Parkersburg 

Special river stations (7 a. m. to 7 a. m.): 

Charleston 

Creston 

Fairmont -. 

Glenville 

Ilinton 

Huntington 

Point Pleasant 

Rowlesburg 

St. Marys 

Wheeling 

Williamson 

Cooperative stations (7 p. m. to 7 p. m.): 

Beckley 

Bens Bun 

Cuba 

Elkhorn 

Grafton 

Ryan 

Wellsburg 



0.00 
1. 94 
0.53 
1.90> 

0.00 
0.00 
0.00 

2.45 
0.90 
1.37 
2.36 
2.00 
0.33 
0.62 
1.03 
0.15 
0.90 
0.51 
T. 
2.50 
1.98 
0.49 
1.29 
1.60 
1.38 
0.75 
0.36 
1.98 
0.62 
2.07 
1.16 



0.00 
0.08 

0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
T. 
0.00 
0.00 

0.00 
0.20 
0.00 
0.00 
0.00 
0.00 
0.23 



2.21 
1.46 
2.14 
1.82 

0.69 
T. 
2.00 

0.85 
1.95 
1.52 
2.00 
1.50 
1.09 
0.30 
2.20 
1.50 
1.23 
2.91 
1.20 
0.72 
1.03 
1.43 
1.77 
1.90 
1.97 
0.90 
1.36 
1.12 
2.13 
1.14 
1.94 



0.00 
0.05 

0.00 
0.00 
0.03 
0.00 
0.00 
0.00 
0.00 
0.00 
0.19 
0.18 
0.00 

0.00 
0.00 
0.30 
0.00 
0.00 
0.00 
0.41 



4.15 
2.66 
2.89 

1.74 

2.57 
0.03 
2.15 

2.50 
5.25 
5.61 
2.64 
2.00 
1.87 
2.70 
3.00 
1.97 
2.86 
3.28 
1.67 
2.80 
4.61 
2.68 
4.45 
4.10 
4.39 
1.90 
1.46 
3.65 
3.12 
1.78 
4.84 



0.22 
0.80 

0.00 

0.00 
0.02 
0.00 
0.00 
0.00 
0.20 
0.00 
0.00 
0.53 
0.00 

0.00 
0.52 
0.28 
0.97 
0.35 
0.36 
0.76 



1.11 
0.91 
1.40 
0.48 

4.06 
2.78 
3.50 

0.12 
1.55 
2.13 
0.24 
0.30 
2.55 
1.35 
1.62 
2.29 
0.97 
1.48 
2.20 
0.20 
0.88 
2.06 
1.41 
1.15 
1.87 
1.35 
2.29 
0.47 
2.25 
0.32 
1.40 



0.49 

1.84 

1.20 
1.60 
1.28 
1.15 
0.62 
2.24 
1.80 
0.90 
1.63 
1.86 
0.78 

0.95 
1.21 
1.48 
2.05 
0.80 
1.40 
1.83 



0.00 
0.25 
0.01 
0.25 

1.22 
1.40 
1.19 

0.55 

0.91 
0.53 
0.30 
0.25 
0.88 
0.75 
0.60 
0.37 
0.85 
0.76 
1.42 
0.94 
0.87 
0.50 
0.41 
0.90 
1.00 
0.70 
0.70 
0.75 
0.54 
0.34 
0.81 



0.70 
0.24 

1.35 
1.47 
0.92 
1.20 
1.74 
1.90 
1.18 
0.90 
1.00 
0.50 
1.70 

1.40 
0.77 
0.95 
0.05 
1.05 
1.29 
0.78 



7.47 
7.22 
6.97 
6.19 

8.54 
4.21 

8.84 

6.47 
10.56 
11.16 

7.54 
6.05 
6.72 
5.72 
8.45 
6.28 
6.81 
8.94 
6.49 
7.16 
9.37 
7.16 
9.33 
9.65 

10.61 
5.60 
6.17 
7.97 
8.66 
5.65 

10.6*5 



1.41 
3.01 

2.55 
3.07 
2.25 
2.35 
2.36 
4.14 
3.18 
1.80 
2.82 
3.07 



2.35 
2.70 
3.01 
3.07 
2.20 
3.05 
4.01 



22 



PRECIPITATION IN OHIO EIVER WATERSHED MARCH 23-27, 1913. 



Table No. 2. — Daily amounts of precipitation {in inches and hundredths) at representative stations in the 
watershed of the Ohio River. Mar. 28-27, 1913 — Continued. 



Mar 23, 



Mar. 24. 



Mar. 25. 



Mar. 26. 



Mar. 27. 



Total. 



KENTUCKY. 



Regular stations (midnight to midnight): 

Lexington 

Louisville 

Special river stations (7 a. m. to 7 a. m.): 

Beattyville 

Burnside 

Catlettsburg 

Falmouth 

Frankfort 

High Bridge 

Maysville 

Paducah 

Pikeville 

Williamsburg 

Cooperative stations (7 p. m. to 7 p. m.): 

Anchorage 

Berea 

Calhoun 

Edmonton 

Franklin 

Hopkinsville 

Irvington 

Marion 

MIddlesboro 

Scott 

Shelby City 



Regular stations (midnight to midnight): 

Evansville 

Fort Wayne 

Indianapolis 

Terre Haute 

Special river stations (7 a. m. to 7 a. m.): 

Attica 

Bluffton 

Elliston 

Madison 

Mount Vernon 

Shoals 

Cooperative stations (7 p. m. to 7 p. m. ): 

Anderson 

Berne 

Butlerville 

Collegeville 

Connersville 

Crawfordsville 

Eminence 

Farmersburg 

Greenfield 

Huntingburg 

Huntington 

Judy ville 

Kokomo..., 

Mauzy 

Moores Hill 

Princeton 

Rome 

Salamonia 

Salem 

South Rend 

Underwood 



0.00 
T. 

0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 

0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
T. 
0.00 



0.29 
2.08 
1.27 
1.05 

0.37 
0.00 
0.00 
0.36 
0.00 
T. 

2.34 
2.30 
0.14 
1.20 
0.68 
2.80 
1.60 
0.78 
1.25 
2.27 
1.80 
1.97 
2.18 
0.56 
0.33 
0.05 
0.00 
3.55 
0.10 
1.15 
0.01 



0.21 
0.15 

0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
T. 
0.00 
0.00 
0.00 

0.20 
0.00 

T. 
0.00 
0.20 
0.20 
0.05 

T. 
0.00 
2.86 
0.00 



0.90 
1.98 
2.76 
2.45 

2.80 
3.80 
1.10 
2.74 
0.21 
0.37 

1.50 
2.34 
2.57 
1.14 
1.85 
2.30 
1.95 
1.92 
2.56 
4.50 
1.05 
1.13 
1.57 
2.25 
1.63 
2.00 
0.10 
1.16 
2.00 
0.53 
2.10 



1.79 
4.95 

0.00 
0.00 
0.00 
0.30 
0.11 
0.05 
0.16 
1.20 
0.00 
0.00 

3.50 
1.20 
1.40 
0.00 
0.00 
0.32 
3.75 
2.28 
0.00 
1.50 
0.61 



4.01 
0.69 

1.56 
0.77 

2.28 
3. CO 
6.10 
3.67 
1.65 



2.51 
2.56 
4.43 
1.86 
5.67 
2.20 
1.45 
2.23 
2.32 
0.52 
1.95 
1.51 
1.97 
5.59 
2.78 
4.37 
3.13 
3.04 
3.10 
0.88 
3.30 



2.46 
0.87 

3.04 
2.25 
2.36 
3.23 
3.35 
3.02 
2.03 
3.64 
0.80 
2.02 

1.72 
5.25 
2.46 
4.30 
4.85 
2.53 
1.80 
1.43 
1.75 
2.38 
4.16 



0.30 
0.40 
0.34 

0.19 

0.00 
0.10 
1.20 
2.27 
2.55 
1.80 

0.50 
0.42 
1.56 
0.20 
1.46 
0.70 
0.25 
0.21 
1.00 
0.00 
0.30 
0.27 
0.00 
0.98 
2.10 
1.05 
2.46 
1.08 
1.10 
0.60 
2.30 



0.01 
T. 

3.28 
3.25 
0.92 
0.96 
1.06 
1.24 
1.29 
0.40 
1.40 
2.05 

0.08 
0.60 
0.13 
0.09 

T. 
0.10 
0.20 

T. 
1.60 
0.23 
0.39 



0.02 
0.21 
0.08 
0.10 

0.63 
0.60 
0.20 
T. 
0.37 
0.45 

0.14 
0.19 
0.57 
0.00 
0.32 
0.00 
0.00 
T. 
0.15 
0.00 
0.30 
0.14 
0.38 
0.27 
0.08 
0.06 
0.09 
0.21 
0.20 
0.18 
0.21 



4.47 
5.97 

6.32 
5.50 
3.28 
4.49 
4.52 
4.31 
3.48 
5.24 
2.20 
4.07 

5.50 
7.05 
3.99 
4.39 
5.05 
3.15 
5.81 
3.70 
3.35 
6.97 
5.16 



5.52 
5.36 
6.01 
4.56 

6.08 
7.50 
8.60 
9.04 
4.78 
9.28 

6.99 
7.81 
9.27 
4.40 
9.98 
8.00 
5.25 
5.14 
7.28 
7.29 
5.40 
5.02 
6.10 
9.65 
6.92 
7.53 
5.78 
9.04 
6.50 
3.34 
7.92 



PRECIPITATION IN OHIO RIVER WATERSHED MARCH 23-27, 1913. 



23 



Table No. 2. — Daily amounts of precipitation (in inches and hundredths) at representative stations in the 
watershed of the Ohio River, Mar. 23-27, 1913 — Continued. 



SOUTHERN ILLINOIS. 

Regular stations (midnight to midnight): 

Cairo 

Special river stations (7 a. m. to 7 a. m.): 

Sha wneeto ivd , 

Chester 

Mount Carmel 

Cooperative stations (7 p. m. to.7 p. m.): 

Albion 

Carbondale 

Carlyle 

Danville 

Equality 

Flora 

Manteno 

Metropolis 

Palestine 

Tuscola 

WESTERN VIRGINIA. 

Regular stations (midnight to midnight): 

Lynchburg 

■VVytheville 

Special river stations (7 a. m. to 7 a. m.): 

Buchanan 

Speers Ferry 

NORTHERN ALABAMA. 

Special river stations (7 a. m. to 7 a. m.): 

Bridgeport 

Florence 

Guntersville 

WESTERN NORTH CAROLINA. 

Regular station (midnight to midnight): 

Asheville 

TENNESSEE. 

Regular stations (midnight to midnight): 

Chattanooga 

Knoxville 

Memphis 

Nashville 

Special river stations (7 a. m. to 7 a. m.): 

Carthage 

Celina 

Clarksville 

Elizabethton 

Johnsonville 

Kingston 

Newport 

New River 

Cooperative stations (7 p. m. to 7 p. m.): 

Ashwood 

Benton _ 

Byrdstown 

Cedar Hill 

Dover 

Erasmus 

Jackson 

Kenton 

McMinnville 

Mountain City 

Rogersville 

Savannah 



Mar. 23. 



0.00 
0.00 
T. 

T. 
0.00 

T. 
2.20 

T. 

T. 
1.34 
0.00 
0.16 
2.05 



0.00 
0.00 



0.00 
0.00 



0.00 
0.00 
0.00 



0.00 



0.00 
0.00 
0.08 
0.00 

0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 

0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 

o.oo 

0.00 
0.00 
0.00 
0.00 



Mar. 24. 



0.02 

0.52 
0.62 
0.86 

2.10 
0.25 
1.60 
1.35 
0.27 
2.38 
0.29 
0.35 
1.34 
1.22 



0.00 
0.00 



0.00 
0.00 



0.00 
0.00 
0.00 



0.01 



0.00 
0.00 
0.00 



0.00 
0.00 
T. 
0.00 
0.10 
0.00 
0.00 
0.00 

0.00 
0.00 
0.00 
0.30 
0.24 
T. 
0.00 
0.24 
0.00 
0.00 
0.00 
0.48 



Mar. 25. 



1.62 
2.80 
6.20 

6.23 
4.24 
1.92 
1.40 
3.02 
3.30 
0.45 
3.74 
3.67 
1.23 



0.00 
0.00 



0.00 
0.00 



0.00 
0.00 
T. 



0.09 



0.10 
0.08 
1.23 
1.11 

0.01 
0.00 
0.53 
0.00 
0.86 
0.00 
0.00 
0.00 

T. 

T. 
0.00 
0.20 
0.91 
0.33 
0.04 
0.48 
0.00 
0.00 
0.00 
0.63 



Mar. 26. 



2.44 
0.22 

1.50 

0.71 
0.52 
0.58 
0.54 
0.97 
0.40 
0.15 
1.25 
0.40 
0.42 



0.27 
1.70 



0.00 
0.26 



0.00 
0.25 
0.10 



1.58 
2.18 
0.56 
1.85 

4.50 
4.20 
3.47 
1.00 
4.00 
0.22 
T. 
1.42 

2.50 
0.53 
0.00 
4.72 
2.67 
2.07 
2.75 
1.79 
1.25 
0.63 
0.33 
5.95 



Mar. 27. 



0.02 

0.74 
0.82 
0.60 

0.07 
0.07 

T. 
0.23 

T. 
0.11 
07 
0.05 
0.00 
0.08 



0.54 
0.88 



2.20 
2.00 



1.86 
1.44 
0.72 



0.40 



0.03 
0.16 
T. 
0.01 

1.90 
2.16 
1.26 
1.60 
0.64 
2.86 
1.75 
2.54 

0.00 
0.75 
3.70 
0.15 

T. 
0.92 

T. 
0.01 
1.85 
1.23 
2.13 
0.04 



Total. 



5.32 
4.46 
9.16 

9.11 
5.08 
4.10 
5.72 
4.26 
6.19 
2.30 
5.39 
5.57 
5.00 



0.81 
2.58 



2.20 
2.26 



1.86 
1.69 
0.82 



1.19 



1.71 
2.42 
1.87 
2.97 

6.41 
6.36 
5.26 
2.60 
5.60 
3.08 
1.75 
3.96 

2.50 

1.28 
3.70 
5.37 
3.82 
3.32 
2.79 
2.52 
3.10 
1.86 
2.46 
7.10 



24 THE FLOODS OF 1913. 

A chart of the aggregate rainfall for the period covered by the table is also presented. This 
chart No. 1 shows by appropriate shading the intensity of the fall in the several parts of the 
watershed. The figures in red express the total rainfall, in inches and tenths, for the stations 
and the period covered by Table No. 2. It should be remembered that rain did not begin to fall 
south of the Ohio River until March 25. High water in the smaller southern tributaries there- 
fore was about a day later than in the northern tributaries. 

A careful analysis of the rainfall statistics of the 1913 flood as published in Table No. 2 
reveals the important fact that in the beginning rain fell over the headwaters of the streams in 
Ohio and Indiana sufficient to fill their bank full. Subsequently rain ceased on the headwaters, 
but continued in the lower reaches of all streams in the States named and extended into the 
watersheds of the tributaries which enter the Ohio River along the left bank. It is also dis- 
closed that practically all of the flood-producing rain fell over the States of Ohio. Indiana, and 
Illinois and that the two headwater tributaries of the Ohio, viz, the Allegheny and the Monon- 
gahela. were not serious factors in the causation of the flood under discussion. 

The rains of the spring of 1912 which caused the greatest flood that the lower Mississippi 
Valley yet has experienced likewise fell not on the headwaters of the Missouri, the upper 
Mississippi, the Wabash, nor the Tennessee and the Cumberland, but on the middle and lower 
portions of the respective basins of those streams, whence the inference that in methods of flood 
protection by reservoirs or otherwise it will be necessary to care not only for the rain which 
falls on the headwaters, but also in the low-lying portions of the respective basins. 

Table No. 3 follows with river-gage readings made at 8 a. m. eastern time, on the tributaries 
of the Ohio in Pennsylvania, Ohio, Indiana, Illinois. Kentucky, Tennessee, and West Virginia. 
The table also contains a column comparing the absolute high water of the March. 1913. 
flood with records of previous high water. It appears that in general the northern tributaries 
in Ohio and Indiana were about as much above previous high water as the southern tributaries 
Jacked of reaching previous high-water stages. (See page 27.) 

The March, J 913, food in the Ohio River. — The flood in the main stream was essentially due 
to the volume of water discharged by the rivers of Ohio and Indiana augmented, of course, by 
the discharge of the southern tributaries. The watershed of the Allegheny did not receive as 
much precipitation as did that of the Beaver., immediately to the westward. Practically no 
rain fell over the Monongahela watershed until the 26th, and while it continued over that 
watershed until the morning of March 27, the river itself was at no time in flood. At some 
places the Allegheny exceeded the high stage of the flood of 1865, but the failure of the 
Monongahela to reach flood stage fortunately prevented a record-breaking rise in the Ohio at 
Pittsburgh. 

At Wheeling, W. Va., 91 miles below Pittsburgh, the volume of the flood was much greater 
than at Pittsburgh, doubtless due to the great amount of water discharged by the Beaver 
River. The highest mark at Wheeling fell 2 feet short of the crest of the 1881 flood. 

The most important tributary of the Ohio between Wheeling and Parkersburg is the 
Muskingum, which enters the Ohio at Marietta, 12 miles above Parkersburg. The precipitation 
over the Muskingum watershed was remarkably heavy. As a result the river crested at Zar.es- 
ville, Ohio, 70 miles from the mouth of the Muskingum, at 51.8 feet, 15 feet higher than any 
previously recorded stage. The immense volume of water carried by that river arrived at its 
junction with the Ohio simultaneously with the flood wave from the Allegheny and the Beaver. 
thus producing higher stages by as much as 5 feet than in the 1884 flood. One of the 
first effects of the flood waters near the mouth of the Muskingum was to isolate Parkers- 
burg and Marietta from communication with the outside world by telegraph or telephone: and 
while some loss resulted from lack of means to transmit warnings, there was still great loss in 
the district to property which could not be protected. 

It is interesting to note the volume of water which flowed past Parkersburg in 24 hours. 
Fortunately the Weather Bureau is able to present hourly river gage readings made under 



THE MARCH, 1913, FLOOD IN THE OHIO KIVEE. 



25 



the supervision of the Parkersburg station from 11 a. m., March 26, to 8 p. m., March 31, as in 
the table below. 





Feet and tenths. 




Feet and tenths. 


26th, 11 a. m 


25.5 


28th, 5 a. m 


53.9 


12 m 


26. 8 


6 a. m 


54.3 


1 p. m 


28. 


7 a. m 


54. 5 


2 p. m 


28. 7 


8 a. m 


54.9 


3 p. m 


29. 9 


9 a. m 


55. 2 


4 p. m 


30. 5 


10 a. m 


55. 5 


op. ra 


31.8 


11a. m 


. .* 55.7 


6 p. m 


32.5 


12 m 


56. 


7 p. m 


33.5 


lp.ra 


56. 1 


8 p. m 


34.5 


2 p. m 


56.4 


9 p. m 


35. 4 


3p.m.... 


56. 5 


10 p. m 


36.2 


4 p. m 


56.7 


11 p. m 


37.0 


sp. 111 


56. 9 


12 md 


37. 8 


6 p. m 


57.1 


27tti, la. m 


38.5 


7 p. m 


57.2 


2 a. m 


39.3 


8p.m 


57.3 


3 a. m. . .. 


40. 1 


9 p. m 


57.5 


4 a. m. . .. 


41.1 


10 p. m 


57. 6 


5a.m.... 


41.6 


11 p. m 


57. 7 


6 a. m. . .. 


42. 


12 md 


57.8 


7 a.m.... 


42. 3 


29th, la. m 


57. 9 


8 a. m. . .. 


43.0 


2 a. m 


58.0 


9 a.m.... 


43. 8 


3 a. m 


58. 1 


10a. m 


44.6 


4 a. m 


58.2 


11 a. m. . .. 


45.1 


5 a. m 


58. 3 


12 m 


45. 8 


6 a. m 


58. 4 


1 p. m 


46. 2 


7 a. m 


58. 5 


•' p. m 


47. 


8 a. m 


58. 5 


3 p. m 


47.7 


9 a. m 


58. 6 


4 p. m 


48.1 


10 a. m 

11 a. m 


58. 6 


S p. m 


48. 8 


58. 7 


6 p. m 


49. 2 


12 m 

1 p. m 


58. 7 


7 p. m 


50. 


58. 8 


8 p. m 


50. 3 


2 p. m 


58. 8 


9 p. m 


50.7 


3 p. m 


58.8 


10 p. m 


51.3 


4 p. m 


58.8 


11 p. m 


51.8 


5 p. m 


58.8 


12 md 


52. 3 


6 p. m 


58.9 


28th, "1 a. m 


52. 8 


7 p. m 

8 p. m 


58.9 


2 a. m 


53.2 


58. 9 


3 a. m 


53. 4 


9 p. m 


58. 8 


4 a. m 


53.7 


10 p. m 


58. 7 



I lowly river gage readings, Parkersburg, W. Va.. 11 a. m. Mar. ,'6-8 p. m. Mar. 31, 1913. 

Feet and tenths. 

29th, 11 p. m 58. 6 

12md 58.5 

30th, 1 a. m 58.4 

2 a. m 58.4 

3 a. m 58. 3 

4 a. m 58.2 

5 a. m 58. 1 

6 a. m 58. 

7 a. m 57.9 

8 a. m 57.9 

9 a. in 57. 8 

lOa.m 57.6 

11a. m 57.6 

12m 57.6 

1 p. m 57.5 

2 p. m 57.4 

3p.m 57.2 

4 p. m 57.0 

5 p. m 56.9 

6p.m 56.7 

7 p. m 56.5 

8 p. m 56.3 

9 p. m 56.1 

10 p. m 55.9 

lip. m 55.7 

12 md 55.5 

31st, la. m 55.3 

2 a. m 55.1 

3 a. m 54.9 

4 a. m 54.7 

5 a. rri 54. 5 

6 a. m 54.3 

7 a. m 54.1 

8 a. m 53.8 

lla.m 53.0 

1 p. m 52.5 

3 p. m 52.0 

4p.m 51.8 

5 p. m 51.5 

6p.m 51.2 

7 p. m 50. 9 

8 p . m 50. 6 

In order to form an idea of the approximate run-off during the storm period we have 
computed the discharge of the Ohio at Parkersburg, W. Va., for the seven days, March 26-April 
1. when the river was above a stage of 35 feet. The rain period in the watershed above Parkers- 
burg was March 24-27. The increased flow in the river became noticeable on March 26, when 
the mean stage, 8 p. m. to 8 p. m., was 22.1 feet. A stage of 35 feet was reached and passed 
at 8 p. m. of the 26th. The total discharge for the seven clays when the river was above that 
stage was approximately 323,654,400,000 cubic feet ; deducting from this amount that wdiich may 
be called the normal flow at the time the rains began, and assuming that the normal flow was 
constant for the seven days, we have 323.654,400.000—27.216.000,000=296.438,400.000 cubic feet 
as the excess discharge due to the rains. 

The rainfall for the watershed of the Ohio above Parkersburg averaged 4.7 inches; in 
some parts of the watershed the actual rainfall was as much as 10 inches, while in other parts 
it was less than 2 inches. The computed rainfall on the basis of an average of 4.7 inches per 
square mile is 406,188,200.000 cubic feet, or about 2.8 cubic miles of water. The approximate run- 
off as determined above was 296,438.400,000 cubic feet, or about 2 cubic miles of water, and the 
percentage of excess run-off to total rainfall for the seven days was 73, an extremely high value 
of run-off, but not an improbable one, considering the time of year and the probable height 
of ground water at the beginning of the period. The above figures of excess discharge for the 
seven days correspond to a discharge of 13 second-feet per square mile of watershed. The 
maximum discharge was, however, 17.5 second- feet per square mile. 

From Parkersburg to Cincinnati the river rose very irregularly. At the latter place it rose 
nearly a foot an hour during the night of the 25th-26th and reached 50 feet, the flood stage, by 
7 a. m. of the 26th. By that hour the river stood 21.5 feet higher at Cincinnati than at Mays- 
ville, Ky., the last named being 60 miles upriver. 

The explanation of this extraordinary 24-hour rise at Cincinnati was furnished by Mr. 
W. C. Devereaux of the Cincinnati office, viz, that it was due to the volume and strength of the 



26 THE FLOODS OF 1913. 

current issuing from the Great Miami River where it joins the Ohio, about 15 miles below 
Cincinnati, thus causing the Ohio to back up in that stretch of the stream between the mouth 
of the Great Miami and Cincinnati. A river packet encountered the output of the Great Miami 
on the night of March 25-26, and so great was the force of the current that the captain of the 
packet was obliged to steer almost directly into the current to prevent being carried onto the 
opposite bank of the riA'er. 1 

The flood discharge of the Scioto came out 24 hours later than that of the Great Miami, as 
did also that of the Muskingum and other small rivers; thus there was not perfect synchronism 
in the flood discharge of the Ohio and Indiana streams. To this lack of synchronism may be 
ascribed the immunity from higher stages than were actually recorded from Cincinnati to 
Cairo. At the time the river reached flood stage at Cincinnati the principal flood wave was 
about 24 hours upriver from Parkersburg. The latter place is 286 miles upstream from Cincin- 
nati, or about three days run for a normal flood wave. As a matter of fact, the crest of the rise 
at Cincinnati was not reached until six clays after the river attained a flood stage. Meantime 
much of the flood water first discharged into the river had passed downstream. The Cincinnati 
crest, 70 feet, was reached at 5 a. m. of April 1, and that height was maintained until midnight 
of that date, when the river began to fall slowly. As a result of the lack of synchronism in 
one; however, the highest water on record was reached at all points from Parkersburg to 
the discharge of flood waters into the trunk stream, the flood wave in the latter was a very flat 
Maysville (see diagram 1). 

The flood, Cincinnati to Louisville. — At Madison, Ind.. 86 miles below Cincinnati, the 
crest of the flood wave, 62.8 feet, was reached on April 1, 1913. That stage exceeds the record 
of the 1884 flood by 1 foot and is the only instance where the 1913 flood exceeded the earlier 
flood between Cincinnati and Louisville. At Louisville the crest stage of 44.9 feet on April 2, 
1913, fell short of previous high record by 1.8 feet. For a detailed account of the flood at 
Louisville see accompanying papers, page 81. 

The flood, Louisville to Cairo. — The river between Louisville and Cairo was about 9 feet 
below flood stage when the rains of March 25-27 set in. As at Cincinnati it rose rather rapidly 
during the 48 hours immediately following the rains, but the bulk of the flood water from the 
Wabash, on the north, did not come in until the night of March 29-30. One effect of the 
Wabash flood was to back up the Ohio, thus causing relatively high-crest stages at Mount 
Vernon, Ind., Henderson, Ky., and Evansville, Ind. The crest at Evansville, 48.3 feet, occurred 
about noon of April 5, and is believed to have been due to a short period of heavy rains over 
the lower Ohio and Green River Valleys, combined with the local ponding effect of the 
Wabash. 

At Cairo, 111., the initial stage on March 25, was 40.9 feet, 4.1 feet below flood stage. Flood 
stage was reached on March 27, and a crest stage of 54.8 feet on April 4. This stage is 0.8 foot 
higher than ever before recorded. The river stood at this unprecedented stage until April 9, 
and then began to fall very slowly. The stages of the river at Cairo are, of course, conditioned 
on the levees withstanding the water, and since those protecting Cairo have been raised 
and strengthened in recent years, we should expect higher stages at Cairo than have hitherto 
been recorded. Fortunately, the Mississippi was not at a flood stage at the mouth of the Ohio, 
and thus the Cairo stage was not so high as it might have been had the conditions of the 1912 
flood prevailed. Nevertheless the river passed slowly above the hitherto record stage of 54 
feet reached on April 6, 1912, and on April 4, 1913, touched the mark of 54.8 feet — higher 
by 0.8 foot than ever before recorded. It hung at the 54.8-foot mark for a period of about 72 
hours; meanwhile the people who had remained at Cairo, assisted by the State militia, were 
making heroic efforts to save the levees and the dty from total destruction. On April 7, 1913, 
the river began to fall, almost imperceptibly at first, but more rapidly by the 18th, and on the 
22d of the month it passed below the flood stage, thus ending the most memorable flood that 
lias yet menaced the lower Ohio River. 

1 The effect of the Great Miami water was noticeable as far up river as Ripley, Ohio. 



DAILY GAGE READINGS, MARCH 22-29, 1913. 



27 



Table No. 3. — Daily river gage readings (in feet and tenths) Mar. 22-29, 1913, and crest stages as compared with previous 
highest water at special river stations in the watershed of the Ohio . 



Flood 
stage 
(feet). 



Previous highest 
stage. 



! Height. 



2.8 
1.3 

2.2 
2.4 
3.5 

3.8 
6.5 
11.9 

3.3 



15. 



LAKE ERIE SYSTEM. 

Sandusky River: 

Tiffin, Ohio 7 18. 5 Apr. 2, 1904 

Fremont.Ohio 10 16.5 - —1904 

Maumee River: 

Fort Wayne, Ind 15 22.5 Mar. 8,1! 

Napoleon, Ohio 13 18. 8 Mar. 2, 1910 

OHIO RIVER SYSTEM. 

Clarion River: 

Clarion, Pa m in 

Conemaugh-Kiskiminetas : 

Johnstown, Pa 7 21.0 

Saltsburg, Pa 6 22.1 

Allegheny River: 

Olean, N.Y j 12 19.3 

Warren, Pa 14 17.4 

Franklin, Pa 15 15.4 

Parker, Pa 20 28. 

Freeport, Pa 20 32.7 

Springdale, Pa 27 33. 

Cheat Eiver: 

Rowlesburg, W. Va 14 22.0 

Youghiogheny River: 

Confluence, Pa 10 18.6 

West Newton, Pa 23 30.6 

Mouongahela River: 

Fairmont, W . Va 25 37.0 

Greensboro, Pa IS 39.0 

Lock No! 4, Pa 28 42.0 

Mahoning River: 

Youngstown, Ohio 5 15.8 

Beaver River: 

Beaver Falls, Pa ! 11 15.4 

Tuscarawas River: , 

Canal Dover, Ohio s 12.0 

Muskingum River: 

Coshocton, Ohio S 22.0 

Zanesville, Ohio 25 36.8 

Beverly, Ohio 25 35.0 

Little Kanawha River: 

Glenville, W. Va 20 21.2 

Creston, W. Va 20 25.8 

New-Great Kanawha River: 

Radford, Va 14 34.0 

Hinton, W. Va 14 23.0 

Charleston, W. Va 30 46.9 

Big Sandy River: 

Williamson, W. Va 26 24.5 

Pikeville, Ky 40 39. 5 

Scioto River: 

Columbus, Ohio 17 21.3 

Cireleville, Ohio 12 19.3 

Chillicothe, Ohio 14 28. 3 Mar. 24, 1898 

Licking River: 

Falmouth, Ky 25 38. 

Miami River: 

Dayton, Ohio 18 21.3 • ,1866 3.0 

Hamilton, Ohio 12 21.2 Mar. 24,1898 3.0 

t Estimated. 

a Dyke broke on 25th and flood waters passed around the gage 
subsequent readings not comparable with preceding ones. 
1 About 22.9. 



Date. 



Mar. 20,1905 

May 31,1889 
, 1859 

June 1,1889 
Mar. —,1865 
Apr. 30,1909 
Mar. —,1865 
Feb. 18,1891 
Mar. 16,1908 

July 10,1888 

Mar. 14,1907 
Feb. 27,1912 

July 10,1888 

....do 

July 11,1888 

Jan. 21,1904 

Jan. 22,1904 



Mar. 24,1898 

....do 

Mar. —1898 

Jan. 9,1907 
Apr. 20,1901 

Sept. 15, 1878 
Sept. 13,1878 
Sept. 29, 1861 

June 14,1907 
Aug. —1903 

Mar. 23,1898 
July 17,1884 
Mar. 24,1898 

, 1854 



March, 1913. 



6 


1 

8.0 
8.8 

0.6 

4.6 



1.0 



23 



6.7 
0.8 



2.6 

1.2 

2.0 
2.2 
3.2 
3.2 
6.2 
11.4 

3.3 

1.3 
1.8 

15.0 
7.9 
8.5 

0.5 

4.4 



1.2 
9.7 
7.6 

1.4 
3.1 

1.2 
3.1 

5.9 

3.4 
5.0 

4.8 



1.6 
4.2 



3.0 
3.0 



24 



7.0 

9.4 



19.6 
9.2 



3.2 

2.5 
1.0 

2.7 
3.8 
6.1 
5.0 
5.9 
11.2 

3.2 

1.3 
1.5 

14.9 

7.8 
8.2 

4.7 

6.6 

2.3 

2.5 
9.9 



1.4 
2.9 

1.1 
3.0 

5.7 

3.2 
4.9 

6.2 

1.6 

4.0 



7.0 

4.8 



12.5 
13.5 



a24.0 
16.0 



2.5 
1.0 

6.6 
8.0 
11.6 
15.4 
16.2 
19.0 

3.2 

1.1 
1.3 

14.8 

7.7 
9.0 

15.5 

13.2 

7.0 

11.0 
21.2 
16.6 

1.6 
2.8 

1.0 
2.9 

5.5 

3.1 
4.8 

21.9 
11.6 
11.9 



*24.0 
19.6 



19. 4f 
21.5 

26.0 
22.0 



12.0 

3.0 
1.6 

14.0 
14.1 
22.0 

24.5 
26.4 
30.0 

3.1 

1.6 
2.2 

14.8 
8.0 
10.0 

•22.9 

16.7 

13.0 



2.0 

4.2 

1.0 
2.8 
5.5 

3.2 
4.5 

20.9 
24.2 



29.1 
28.1 



16. Of 
21.5 

26.0 
25.0 



12.3 

5.5 
6.2 

14.4 
14.8 
21.1 
24.5 
31.9 
36.5 



4.9 
7.4 

20.2 

14.6 
16.2 



17.4 
15.0 



51.8 
46.5 

15.2 
16.0 

7.3 
6.5 
10.2 

26.4 
34.0 

19.7 
20.3 



33.8 



522.2 
25.0 



12. Of 
14.3 



25.1 
22.5 



10.2 

4.6 
6.3 

15.2 
14.1 
19.5 
23.0 
29.5 
34.3 



4.8 
8.5 

22.4 

18.7 
25.2 



15.1 
16.1 



11.0 
18.9 

9.8 
11.6 
33.0 

18.0 
21.5 

17.4 
16.2 



32.2 



515.7 
19.2 



8. Of 
11.0 

23.7 
18.0 



7.2 

4.0 
4.2 

12.6 
12.5 
15.3 
17.0 
23.5 
28.3 



3.5 

5.7 

19.1 
13.6 
20.2 

10.4 

12.0 

9.0 



3.2 
9.5 

4.2 

7.2 

30.0 

8.9 
10.4 

14.7 
13.8 
24.6 

23.6 

11.6 
14.8 



High 

est. 



19.4 
21.5 



26.1 
25.0 



13.9 

5.5 
7.7 

15.6 
15.2 
22.5 
25.7 
32.2 
36.5 



5.6 
9.7 

23.6 
18.7 
25.2 

'22.9 

17.4 
16.1 

'20.0 

51.8 
46.5 

20.0 
20.4 

11.5 
14.5 

34.8 

30.4 
39.0 

22.9 
24.2 
37.8 

34.1 

29.0 
34.6 



Date. 



Mar. 26 
..do.... 



..do.... 
Mar. 27 



Mar. 

Mar. 
...do. 

...do. 
...do. 

Mar. 
...do. 

Mar. 
...do. 

Mar. 

Mar. 
...do. 

...do. 

Mar. 
...do. 

Mar. 

Mar. 

Mar. 

Mar. 
Mar. 
...do. 

...do. 

Mar. 

Mar. 
Mar. 
...do. 

Mar. 
...do. 

Mar. 
Mar. 
...do. 



28 



Mar. 27 



Mar. 
Mar. 



Com- 
pared 
with 
pre- 
vious 
highest. 



+ 0.9 
+ 5.0 



+ 3.6 

+ 6.2 



- 2.1 

-15.5 

-14.4 

+ 3.7 

- 2.2 
+ 7.1 

- 2.3 

- 0.5 
+ 3.5 

-14.3 

-13.0 
-20.9 

-13.4 
-20.3 
-16.8 

+ 7.1 

+ 2.0 

+ 4.1 

- 2.0 
+ 15.0 
+ 11.5 

- 1.2 

- 5.4 

-22.5 

- 8.5 
-12.1 

+ 5.9 

- 0.5 

+ 1.6 

+ 4.9 
+ 9.5 

- 3.9 

+ 7.7 
+13.4 



2 No record after the 26th. 

3 Obtained by survey. 
< Approximated. 

5 Measurements made at Viaduct Bridge. 



28 



THE FLOODS OF 1913. 



Table No. 3. — Daily river gage readings (in feet and tenths) Mar. 22-29. 1913, and crest stages as compared with previous 
highest water at special river stations in the watershed of the Ohio — Continued. 



Flood 
stage 
(feet). 



Previous highest 
stage. 



Height, 



Date. 



March, 1913. 



24 



26 



27 



1913 



High 

est. 



Date. 



Com- 
pared 

with 
pre- 
vious 
highest 



Little Miami River: 

Kings Mills, Ohio 

Kentucky River: 

Beattyville, Ky 

High Bridge, Ky 

Frankfort, Ky 

White River: 

Anderson, Ind 

Indianapolis, Ind 

Elliston, Ind 

Shoals, Ind 

New River: 

New River, Term 

Wabash River: 

Blufiton, Ind 

Logansport, Ind 

Attica, Ind 

Terre Haute, Ind 

Mount Carmel, 111 

Cumberland River: 

Burnside, Ky 

Carthage, Tenn 

Nashville, Tenn 

Clarksville, Tenn 

Clinch River: 

Speers Ferry, Ya 

Clinton, Tenn 

Holston River: 

Rogersville, Tenn 

French Broad River: 

Dandridge, Tenn 

Hiwassee River: 

Charleston, Tenn 

Little Tennessee River: 

McGhee, Tenn 

Tennessee River: 

Knoxville, Term 

Chattanooga, Tenn 

Bridgeport, Ala 

Florence, Ala 

Johnsonville, Tenn 

Ohio River: 

Pittsburgh, Pa 

Wheeling, W. Va 

Parkersburg, W. Va... 

Point Pleasant, W. Va 

Huntington, W. Va . . . 

Catlettsburg, Ky 

Portsmouth, Ohio 

Maysville, Ky 

Cincinnati, Ohio 

Madison, Ind 

Louisville, Ky 

Evansville, Ind 

Shawneetown, 111 

Paducah, Ky 

Cairo, 111 



3.3 17.8 



37.5 
30.0 

44.0 

18.8 
19.5 
29.6 
34.1 

33.0 

16.7 
17.3 
29.7 
27.7 
28.3 

65.0 
54.3 
55.3 
60.6 

26.6 
45.0 

17.5 

28.0 

32.2 

39.0 

39. 
58.6 
41.0 
32.5 
48.0 

35. 5 
53.1 
53. 9 
60.0 
65. 6 
66.2 
66.3 
65.7 
71.1 
61.8 
46.7 
48.0 
56.4 
54.2 
54.0 



Mar. 1, 1903 

Jan. 30, 1902 

Feb. —1878 

Mar. 23,1904 

Apr. 1,1904 

Mar. 5,1897 

Mar. 30,1904 

Feb. —,1903 

Apr. —,1904 

Feb. —1883 

Aug. 3, 1875 

Feb. 18,1883 

Aug. 7,1885 

Mar. 30,1902 

Apr. 7, 1886 

Jan. 22.1882 

Jan. — , 1882 

Feb. 28,1902 

Mar. 31,1886 

Jan. 23,1906 

May 21,1901 

Mar. 31,1886 

Mar. —,1867 

Mar. — ,1875 

Mar. 11,1867 

, 1867 

Mar. 19,1897 

Mar. 24,1897 

Mar. 15,1907 

Feb. 7, 1884 

Feb. 9, 1884 

Feb. —,1884 

Feb. 9, 1884 

Feb. 12.1884 

....do 

Feb. 14,1884 

....do 

Feb. 15,1884 

....do 

Feb. 19,1884 

Feb. 24,1884 

Feb. 23,1884 

Apr. 6,1912 



0.9 
11.5 
8.6 

4.3 
4.7 



0.8 
11.4 

8.7 

3.8 



7.4 

3.2 

3.2 
3.6 
6.2 
7.1 
11.9 

8.7 
15.8 
21.5 
29.7 

2.6 
9.4 

4.0 

4.9 



8.0 

2.9 

2.5 
3.8 
6.8 
7.0 
13.4 

10.5 
16.4 
17.4 
24.4 

2.4 

8.7 

3.9 
3.5 



10.0 6.4 



8.8 6.8 



4.5 
12.9 
11.2 
18..0 
27.5 

5.3 
8.8 
10.5 
14.1 
19.7 
19.5 
21.5 
23.1 
27.5 
25.1 
11.3 
28.4 
29.3 
33.6 
39.0 



4.8 
12.9 

7.0 
16.0 
28.0 

4.8 
8.3 
10.0 
12.3 
17.9 
17.3 
19.1 
20.3 
24.7 
23.6 
10.8 
28.3 
29.6 
34.3 
39.9 



1.6 
11.3 
8.5 

11.8 

11.0 

11.8 

8.8 

2.6 

12.3 
12.1 
15.9 
14.5 
13.6 

9.1 
15.7 
17.5 
20.6 

2.0 
8.2 

3.6 

3.1 

5.2 

6.2 

4.2 
12.3 
10.4 
13.7 
28.5 

4.5 
7.5 
9.5 
11.1 
16.1 
13.8 
17.4 
18.4 
22.6 
21.6 
10.0 
27.5 
29.3 
34.4 
40.3 



2.0 

11.1 
8.5 

17.6 
18.0 
23.8 
21.6 

2.4 



24.6 
19.5 
18.3 

8.2 
15.5 
16.2 
20.1 

1.6 
7.8 

3.4 

3.0 

4.8 

5.9 

3.5 
11.2 

9.4 
12.0 
28.4 



11.5 
10.0 

10.1 
14.1 
15.5 
16.1 
17.1 
29.3 
27.5 
11.4 
26.0 
28.9 
34.3 
40.9 



33. 7 

11.6 
21.0 
15.8 

20.6 



37.0 
34.6 
35.2 

14.0 



38.6 
33.4 
38.3 

10.2 



27.8 
29.5 

4.1 

20.0 
22. 5f 
31.6 
27.0 
21.4 

15.2 
21.0 
25.0 
31.6 

1.0 
7.6 

3.3 

2.9 

5.2 

5.9 

3.2 
10.1 

8.5 
10.7 
29.4 

20.1 
30.5 
22.1 
13.2 
17.2 
17.2 
21.9 
28.8 
50.3 
43.5 
22.5 
30.1 
31.1 
36.9 
43.5 



31.3 
37.0 



23.5 



19.0 



33.4 
31.2 
23.0 

57.2 
37.5 
39.3 

47.3 

12.0 
23.4 

7.4 

8.5 

13.2 

13.8 

7.3 
13.3 

9.2 
13.7 
32.1 

28.1 
45.5 
43.0 
34.1 
39.4 
41.1 
51.0 
44.3 
57.2 
53.6 
33.6 
36.6 
34.9 
38.5 
45.5 



30.4 
42.2 



5.6 



13.8 



31.6 
30.8 
24.8 

58.0 
43.9 
42.7 
50.5 

17.5 
26.5 

19.1 

12.2 

20.0 

17.5 

20.9 
25.4 
16.4 
14.0 
33.0 

30.4 
50.8 
54.9 
50.6 
54.5 
56.8 
61.9 
57.5 
62.6 
57.0 
38.4 
40.4 
3S.3 
40.7 
47.4 



19.5 
33.5 
37.5 



41.7 



3.9 



12.3 



28.5 
29.2 
27.8 

40.0 
47.0 
42.8 
50.5 

7.2 
29.9 

8.5 

8.4 

14.5 

11.2 

20.1 
31.2 
20.6 
15.7 
33.3 

24.8 
50.0 
58.7 
60.2 
63.8 
65.1 
62.8 
62.8 
66.0 
59.6 
41.1 
43.0 
41.9 
42.6 
49.1 



39.9 
34.6 
38.3 

22.1 
25.7 
31.3 

42.2 

23.5 

20.0 
22.5 
33.4 
31.3 
31.0 

61.5 
47.0 
44.9 
50.9 

18.2 
30.2 

19.1 

16.0 

20.3 

21.6 

21.6 
33.3 
23.7 
18.5 
33.3 

30.4 
51.1 
58.9 
62.8 
66.4 
67.7 
67.9 
66.4 
70.0 
62.8 
44.9 
48.4 
59.5 
54.3 
54.8 



Mar. 26 

Mar. 28 
Mar. 27 
Mar. 28 

Mar. 25 
Mar. 26 
Mar. 27 
Mar. 28 

Mar. 27 

Mar. 26 
..do.... 
Mar. 27 
..do.... 
Mar. 30 

Mar. 28 
Mar. 29 
Apr. 2 
Mar. 28 

Mar. 27 
Mar. 29 

Mar. 28 

..do.... 

..do 



Mar. 27 

Mar. 28 

Mar. 30 

Mar. 31 

Mar. 21 

Mar. 29 

Mar. 2S 
..do.... 
Mar. 29 
Mar. 30 
..do..... 
Mar. 31 

..do 

..do 

Apr. 1 

..do 

Apr. 2 
Apr. 5 

..do 

Apr. 7 
Apr. 4, 7 



+ 2.4 
+ 4.6 

- 5.7 

+ 3.3 
+ 6.2 
+ 1.7 
+ 8.1 

- 9.5 

+ 3.3 
+ 5.2 
+ 3.7 
+ 3.6 

+ 2.7 

- 3.5 

- 7.3 
-10.4 

- 9.7 

- 8.4 
-14. S 

+ 1.6 

-12.0 

-11.9 

-17.4 

-17.4 
-25.3 
-17.3 
-14.0 

-14.7 

- 5.1 

- 2.0 
+ 5.0 
+ 2.8 
+ 0.8 
+ 1.5 
+ 1.3 
+ 0.7 

- 1.1 
+ 1.0 

- 1.8 
+ 0.4 
+ 3.1 
+ 0.1 
+ 0.8 



t Determined by survey. 



COMPARISON WITH PREVIOUS FLOODS. 



29 



THE FLOOD IN THE OHIO RIVER AS COMPARED WITH PREVIOUS FLOODS. 

Fragmentary records of previous floods in the Ohio River at Pittsburgh go back to 1806, 
but probably the best authenticated record of an early flood is that of February 10, 1832, when 
a stage of 35 feet, on the Pittsburgh gage, or 13 feet above the present flood stage, was reached. 
This high record, moreover, stood as such for three-quarters of a century, or until March 15, 
1907. when a stage of 35.5 feet was reached, half a foot higher than the 1832 record. The flood 
at Pittsburgh during March, 1913, came mostly from the Allegheny River, and was not so 
great by about a foot as the January flood of the present year. (See Table No. 1.) 

The greatest of floods in the Ohio River between Pittsburgh and Cairo during the forty- 
odd years that the Weather Bureau has been making daily observations of stages of the river, 
whether the order of magnitude of the flood be determined by the crest stage reached or the 
volume of water carried, was undoubtedly that of February 6-23, 1884, although March, 1913, 
was a close second, and indeed in several stretches of the river as before stated the crest stages 
of the March-April, 1913. flood overtopped those of the earlier flood. From St. Marys, W. Va., 
to probably a short distance below Maysville, Ky.. the stages of the March, 1913, flood were 
the highest ever known; at Parkersburg the previous high record of 1884 was exceeded, by 
5 feet. The Muskingum River of Ohio empties into the main stream at Marietta, Ohio, about 
12 miles upriver from Parkersburg. This river, at Zanesville, Ohio, 70 miles from its mouth, 
attained a stage 15 feet above previously known high water. The result of suddenly discharg- 
ing an amount of water into a stream greatly in excess of the stream's capacity to carry it off 
must be a local ponding in the vicinity of the discharge of the lesser into the greater stream. 
The simultaneous arrival of the headwaters flood wave and the Muskingum's output doubt- 
less explains the extraordinarily high stages at Marietta and Parkersburg. 

Below Parkersburg the excess of the 1913 flood over that of 1884 gradually decreased, due 
to the fact that the northern tributaries had run out and there were practically no floods in the 
southern tributaries ; giving the Ohio flood waters a chance to back up the valleys and spread 
over a large area of country. This was especially noticeable at Point Pleasant, where the water 
spread over the lower portion of the Great Kanawha Valley and the excess in height decreased to 
2 feet, and again below Maysville, where the Ohio Vallej^ widens out and the excess entirely 
disappears. 

In 1884 the flood in the Ohio at Cincinnati must have been augmented by flood waters in 
the tributaries which enter the river near that city to produce the stage it did. 

In 1913 the flood was augmented below Louisville bj r additional water from the Wabash, 
due to heavy opportune rains in that valley; below the mouth of the Green River the 1913 
flood exceeded that of 1884. 

The small table below presents comparative statistics of the floods of February, 1884, and 
March, 1913, respectively, and we also present a diagram which graphically shows the crest 
stages of the floods along the Ohio River from Pittsburgh to Cairo (Diagram No. I). The 
crest stages are plotted in terms of excess in feet above the flood stage for each point at which 
daily gage readings were made. 



Comparative stages, Ohio River, in floods of February, 1884, and March, 1913, respectively. 
[Plus sign indicates a higher stage in 1913 than in 1884, and minus sign the contrary.] 



Stations. 



Pittsburgh, Pa 

Wheeling, W.Va 

Parkersburg, W. Va . . . 
Point Pleasant, W. Va. 
Portsmouth, Ohio 



Crest stage. 



1884 



Feet. 
33.3 
52.4 
53.9 
60.8 
66.2 



1913 



Feet. 
30.4 
51.1 
58.9 
62.8 
67.9 



Differ- 
ence. 



Feet. 
-2.9 
-1.3 
+5.0 
+2.0 
+1.7 



Stations. 



Cincinnati, Ohio 
Louis ville, Ky. . . 
Evansville, Ind.. 

Paducah, Ky 

Cairo, 111 



Crest stage. 



1884 



Feet. 
71.1 
46.6 
48.0 
54.2 
51.8 



1913 



Feet. 
70.0 
44.8 
48.4 
54.3 
54.8 



Differ- 
ence. 



Feet. 
-1.1 

-1.8 
+0.4 
+0.1 
+3.0 



30 



THE FLOODS OF 1913. 



Further comparisons of 1884 a nd 1913 floods. — The antecedent conditions of the two floods 
were quite unlike; the 1913 flood was the result of torrential rains crowded into a very short 
space of time when the conditions were excellent for a large surface run-off, and it came later 
in the year when the soil was free of frost. The 1884 flood was the result of only moderately 
heavy rains (see Chart No. 4 and Table 4) coming on the crest of a midwinter thaw. In fact 
there were two distinct thaws in a period of mild and rainy weather that continued from 
February 4 to February 14. The first and most pronounced thaw culminated on February 5, 
with average temperatures of about 60° F., in the early morning throughout the watershed. 

The period of moderately heavy rains was approximately coincident with the culmination 
of the thaw just mentioned, viz, February 5, G, and 7. The greatest daily fall was but 0.93 
inch for the entire watershed as determined from the reports which appear in Table No. 4. 
A second maximum of rainfall occurred on the 13th at a time when the flood crest was about 
at Cincinnati. The unusually high stage recorded at that place may have been due to the 
effect of this second maximum of precipitation occurring when the river was already at a 
high stage. 



k. 

25 
ZO 
IS 
10 

J 






§1 


I* 




(V o 




*1 


3^ 




1^ 




<5 

<5s<o 

f 










II 
|1 




1 

3 














N \ 








.. 


.^ 










1 


A 














/ / 




*' 


\ > 










^/ 


\ 


^y'i 


~^\ 


S 




y*C. 




\ 














y 






\. 


y 




















vS 


^fl-*^ 






\ 
































1913 CREST STAGES — 
1884 ■■ } — 


-FULL LINE. 
-DOTTED LINE. 


**■' *«. 






















Cr/i/-*" 






































rLG 




jJA 




















j*trx 



Diagram I. — Comparison of crest stages in the Ohio River in floods of 1884 and 1913. 
Temperatures in degrees Fahrenheit in the Ohio Valley at 7 a. m. Washington mean time, Feb. J/-15, 1884. 



Stations. 


Dates. 


1 


2 


3 


4 


5 


6 


7 


8 


9 

53 

45 
48 
36 
59 
40 
58 


10 

36 

34 
36 
30 
35 
34 
43 


11 

43 
36 
45 
35 
43 
42 
50 


12 

52 
40 
48 
42 
53 
58 
57 


13 

58 
52 
55 
38 
47 
38 
57 


14 

40 
28 
20 

8 
19 
15 
27 


15 


Pittsburgh, Pa 


23 
34 
30 
18 
29 
31 
32 


35 
33 
42 
36 

42 
39 
37 


27 
33 
36 
40 
41 
52 
46 


39 

36 
44 
38 
49 
59 
60 


60 
51 
62 
57 
64 
63 
63 


39 
38 
46 
37 
53 
49 
64 


38 
36 
40 
34 
39 
34 
46 


42 
38 
43 
38 
46 
43 
46 


19 


Columbus, Ohio 


20 


Cincinnati, Ohio 


20 


Indianapolis, Ind 


18 


Louisville, Ky 


20 


Cairo.Ill 


24 


Nashville, Term 


18 






Mean 


29 


38 


39 


46 


60 


47 


38 


42 


4S 


35 


42 


50 


49 


23 


20 







The temperatures given in the above table are all from places situated in the lowlands, but 
the corresponding temperatures even at the highest points in the watershed and on the north 
slopes must have been at least 10 degrees above the freezing point. The relatively high tem- 
peratures and the fact that in the month previous to the flood there was an average of about 
2 feet of snow in Ohio, West Virginia, and western Pennsylvania must enter very largely into 
any consideration of the cause of the 1884 flood. Unfortunately, the snowfall records of 30 
years ago are very deficient as to number and unsatisfactory as to quality, but enough is known 
to state that at the time of the thaw there was practically no snow on the ground immediately 



DAILY PRECIPITATION FEBRUARY 4-14, 1884. 



31 



along the Ohio River in Ohio. Indiana, and Kentucky, and perhaps for a distance of 100 miles 
back from the river. There was still some snow on the ground in northern Ohio, West Virginia, 
and western Pennsylvania, and doubtless a very considerable amount in the higher altitudes of 
those regions. There are, however, no definite records of snowfall available for the confirmation 
of this view. Gagings of tributary streams were not made at points where such measurements 
would have been useful in determining the run-off from areas whence no meteorological observa- 
tions are available. We conclude that there must have been a large run-off from melting snow 
from the following facts: (1) The rainfall for the entire flood period, viz, February 4 to 14, 
inclusive, does not seem to have been sufficient to produce the volume of water carried by the 
river. (2) The known horizontal distribution of precipitation will not account for the high 
water in the upper reaches of the Ohio ; that is to say, the region of heavy precipitation was 
mostly over the immediate Ohio Valley below Cincinnati. (3) The fact that the Ohio was at 
flood stage at Cincinnati as early as February 4, the day. that the rain began. This last fact 
in itself clearly indicates that a large quantity of snow water had already entered the stream. 
In the 1913 flood the Ohio at Cincinnati was at a stage of 22.6 feet on the second day of the 
rain period, or at an initial stage of about 17 feet lower than at a corresponding period in the 
1884 flood. 

If the Ohio River on March 24, 1913, had been at, say, a 30-foot stage, which is not unusual 
for March, a new record of high water would have been written for the Ohio Valley. In this 
connection it should be kept in mind that torrential rains and severe winter weather are in- 
compatible, and that in the season when heavy rains are probable the rivers and small streams 
are generally low. 

The two floods here considered, so far as known by observation and experience, represent 
the two diametrically opposite weather conditions under which destructive floods in the Ohio 
are likely to occur. 

Table 4. — Daily precipitation {in inches and hundredths), Feb. h-lh, 188-' f , Ohio watershed. 



Stations. 



Pittsburgh, Pa 

Wellsburg, W. Va 

Helvetia, W. Va 

Marietta, Ohio 

Quaker City, Ohio 

Warren, Ohio 

Ironton, Ohio 

Lebanon, Ohio 

Washington C. H., Ohio 

Cincinnati, Ohio 

Waverly , Ohio 

Dayton, Ohio 

Logan, Ohio 

Columbus, Ohio 

Levering, Ohio 

Canton, Ohio 

Granville, Ohio 

Westerville, Ohio , 

Wooster, Ohio 

Sidney, Ohio 

North Lewisburg, Ohio 

Cleveland, Ohio 

Oberlin, Ohio 

Wauseon, Ohio 

Junction (Paulding County), Ohio 

Sandusky, Ohio 

Upper Sandusky, Ohio 

Toledo, Ohio : 

Vevay, Ind 

Jefiersonville, Ind 



0.45 



0.71 
0.85 
0.69 
0.00 
0.54 
0.15 



1.35 

0.58 

0.55 

T. 

0.58 

0.92 

0.52 

0.27 

0.53 



0.26 
0.45 
0.31 
0.20 
0.22 
0.24 
0.20 
0.00 
0.04 



0.84 



0.76 

1.50 

0.06 

T. 

1.80 

1.70 

0.37 

1.95 



1.56 
0.48 
0.66 
1.28 
0.55 
0.00 
0.33 
0.60 
1.11 



0.45 
1.22 
1.18 
1.27 
0.86 
1.04 
2.02 
0.90 
1.55 
(') 
Probably 



0.80 
1.00 
0.75 
1.41 
1.95 
0.92 
0.53 
1.25 



1.65 

1.74 
1.42 
2.37 
1.40 
1.55 
1.07 
1.95 
0.31 
1.87 
1.25 
1.25 
0.63 
0.61 
0.32 
0.36 
0.65 
0.70 
0.28 
3.50 
(') 
included 



0.03 
0.03 
0.05 



0.01 



0.33 

0.00 

0.00 

0.00 

0.00 

T. 

0.00 

T. 

T. 

0.12 

T. 



0.34 

0.25 

0.31 

0.11 

0.22 

0.00 

0.69 

0.10 

3.74 

0.23 

0.69 

0.00 

0.63 

0.07 

0.00 

0.03 

0.00 

0.02 

0.00 

0. 04 0. 11 

0. 00 0. 10 

0. 01 0. 04 
0.07 
0.03 
0.08 
0.06 
0.00 

0. 00 0. 07 

0. 20 0. 00 

3. 98 0. 00 

in the following 



0.04 



0.50 
0.02 
0.00 
0.09 



0.58 
0.19 
0.16 
0.03 
0.06 
0.14 
0.10 
0.06 
0.06 
0.17 
0.04 
0.07 
0.00 
0.12 
0.10 
0.03 
0.34 
0.22 
0.20 
0.51 
0.28 
0.22 
0.27 
0.30 
0.41 
0.08 
0.00 
0.18 
day. 



0.13 



0.62 

T. 

0.26 

0.00 

0.66 

0.00 



0.14 

T. 

0.02 

T. 

0.02 

0.00 

0.00 

0.00 

0.19 

0.00 

0.01 

0.10 

0.00 

0.00 

0.00 

0.00 

0.00 

0.00 

0.00 

0.07 



0.53 



0.09 

0.89 

0.20 

T. 

0.54 

0.30 



0.59 
0.91 
0.16 
0.45 
0.26 
0.45 
0.15 
0.20 
0.24 



0.14 
0.10 
0.45 
0.24 
0.34 
0.48 
0.83 



0.10 
0.00 
1.13 



0.06 



0.00 

0.00 

T. 

0.58 

0.00 

0.27 



0.06 
0.00 
0.33 
0.00 
0.06 
0.00 
0.07 
0.00 
0.26 



0.20 
0.25 
0.15 
0.46 
0.77 
0.45 
0.32 



0.37 
0.90 
0.39 



0.14 
0.75 
0.14 
0.28 
0.99 
0.67 
0.68 
0.60 



1.18 

0.19 
0.90 
0.46 
0.55 
1.02 
0.60 
0.30 
0.51 
0.71 
0.59 
0.55 
0.57 
0.55 
0.18 
0.23 
0.42 
0.88 
0.19 
0.00 
T. 



0.18 

0.25 

0.74 

0.56 

0.00 

T. 

0.39 

0.50 

1.53 

0.00 

0.45 

0.02 

0.40 

0.02 

0.00 

0.47 

0.70 

0.00 

0.00 



0.25 
0.02 



0.00 
00 
0.04 
0.00 
0.00 
0.62 



Total. 



3.44 

3.75 
4.33 
4.29 
6.27 
3.90 
4.46 
5.26 
5.37 
6.82 
5.10 
4.25 
5.63 
3.58 
4.44 
3.38 
4.12 
3.29 
2.92 
2.S2 
3.70 
3.91 
3.59 
3.38 
3.00 
3.91 
4.01 
2.03 
6.84 
6.52 



32 



THE FLOODS OF 1913. 
Table 4. — Daily precipitation, Feb. .'i-J'i. 188.'i, Ohio watershed — Continued. 



Stations. 



Laconia, Ind 

Suninan. Ind 

Indianapolis. Ind 

La Fayette, Ind 

Logansport. Ind 

Rising Sun, Ind 

Terte Haute, Ind 

Wabash, Ind 

Evansville, Ind 

Louisville, Ky 

Frankfort, Ky 

Bowling Green, Ky.. 

Cairo, 111 

Springfield, 111 

St. Louis, Mo 

Memphis, Tenn 

Dyersburg, Tenn 

Grief, Term 

Gadsden, Tenn 

Bolivar, Term 

Milan, Tenn 

McKenzie, Tenn 

Nashville, Tenn 

Chattanooga, Tenn. . 

Knoxville, Tenn 

Jonesboro, Tenn 

Greenville, Tenn 

Mary ville, Tenn 

Andersonville, Tenn. 

Cary ville, Tenn 

Parksville, Tenn 

Ashwood, Tenn 

Austin, Tenn 



0.90 



0.43 

(') 

0) 

(') 
(') 
(') 

0.00 

0.89 

1.23 

(')' 

0.07 

0.57 

1.32 

T. 

(') 

0.00 

0.00 

0.00 

0.10 

0.57 

0.04 

0.00 

0.22 

0.30 

T. 

0.10 

0.00 

0.00 



Mean. 



0.25 



0.29 



0.00 

2.00 

0.81 

1.18 

1.02 

(') 

2.03 

1.10 

0.50 

2.38 

0.40 

(') 

1.17 

0.37 

0.40 

0.57 

(') 

0) 

0.72 

0) 

1.71 

1.40 

T. 

0.00 

0.00 

0.00 

T. 

0.00 

0.00 

0.00 



0.80 
0.00 



0.69 



6 



0.00 
1.38 
0.63 
0.00 
0.00 

C) 

0.83 

0.35 

0.60 

1.73 

1.91 

2.73 

1.24 

0.03 

0.37 

2.41 

3.42 

(') 

2.30 

(') 

0.70 

1.10 

1.73 

0.04 

0.01 

0.01 

0.00 

0.00 

0.00 

0.00 

T. 

1.20 

0.50 



0.93 



4.11 

0.00 
0.04 
0.35 
0.00 
2. ,85 
0.00 
0.00 
1.50 
0.63 
0.70 
0.00 
0.18 
0.03 
0.02 
0.53 



3.00 

0.52 

2.02 

0.44 

0.10 

0.68 

3.19 

1.53 

0.85 

T. 

1.92 

1.55 

1.90 

3.00 

0.50 

1.45 



0.72 



0.00 
0.00 
09 
0.26 
0.33 
0.00 
0.00 
0.01 
0.27 
0.00 
0.00 
0.00 
0.25 
0.16 
0.24 
1.53 



0.00 

0.57 

0.00 

0.95 

0.70 

0.74 

1.21 

0.78 

0.60 

T. 

2.86 

1.76 

1.26 

1.80 

0.70 

0.00 



0.14 

0. 25 

0.31 

T. 

0.00 

0.25 

0.59 

0.20 

0.60 

0.14 

0.10 

0.00 

0.03 

0.12 

0.03 

0.56 

0.80 

2.12 

0.92 

1.67 

0.70 

0.00 

0.44 

0.52 

1.97 

0.65 

1.10 

0.31 

1.09 

1.36 

T. 

0.20 

0.00 



0.30 



0.35 



10 



0.00 
0.00 
0.04 
0.00 
0.00 
0.00 
0.00 
0.00 
0.00 
0.61 
0.00 
0.00 
0.51 
0.10 
0.08 
0.59 
1.90 
0.00 
0.86 
0.42 
0.25 
0.75 
0.25 
0.12 
0.33 
0.08 
1.30 
0.64 
0.26 
0.00 
0.34 
0.00 
1.40 



1.00 
0.32 
0.08 

(') 
0) 

0.40 

0.17 
0.00 
0.47 
0.60 
0.97 
0.90 
0.11 
0.40 
0.25 
0.05 
0.25 
0.00 
0.07 
0.28 
0.02 
0.25 
0.01 
0.01 
0.02 
0.70 
0.40 
0.00 
0.00 
0.00 
0.00 
1.00 
0.00 



0.21 



0.29 



0.41 

1.15 

0.77 

1.39 

0.75 

0.00 

0.02 

0.00 

0.30 

0.27 

0.03 

0.00 

0.13 

1.14 

1.15 

0.42 

0.26 

0.00 

0.10 

0.00 

0.47 

0.30 

0.53 

T. 

0.00 

0.00 

0.00 

0.00 

0.12 

0.00 

T. 

0.70 

1.20 



0.28 



0.55 
0.38 
0.17 
0.43 
0.65 
0.35 
0.00 
1.36 
0.10 
0.77 
1.23 
0.69 
0.-43 
0140 
0.16 
0.37 
0.34 
1.57 
0.53 
0.49 
0.25 
0.00 
1.08 
1.23 
0.61 
0.17 
0.50 
2.23 
0.29 
1.90 
1.30 
0.10 
0.70 



0.57 



0.00 



0.00 
0.00 
0.00 



0. 17 

T. 

0.00 

0.44 

0.01 

0.00 

0.00 

0.00 

0.00 

0.00 

0.00 

0.00 

0.00 

0.00 

0.00 

0.05 

0.38 

0.98 

0.90 

0.00 

1.29 

0.00 

0.00 

0.00 

0.00 



0.19 



Total. 



7.11 
5.48 
3.37 
3.61 
2.75 
3.85 
3.64 
3.02 
4.81 
8.02 
6.59 
4.76 
4.13 
3.32 
4.02 
7.03 



6.59 
4.88 
5.59 
5.17 
5.50 
6.37 
5.85 
4.34 
4.20 
8.06 
6.36 
6.42 
6.44 
5.20 
5.50 



4.82 



1 Probably Included in the following day. 

The 1884 flood is typical of the midwinter flood that results from moderately heavy pre- 
cipitation in conjunction with the melting of snow and the breaking up of ice in the rivers. 
It seems remarkable that so few midwinter floods have occurred in the past. 

The great flood of 1832, so far as can be ascertained, was a midwinter flood; and these two, 
1832 and 1884. seem to have been the only typical midwinter floods of record in the 81 years 
which have passed since 1832. Other midwinter floods are produced by continued heavy pre- 
cipitation during an open winter, as in January, 1913 (see Chart No. 2), in which the accumu- 
lated rainfall in the Ohio watershed January 1-13, 1913, is shown. The origin of the great 
majority of floods in the lower Mississippi can be traced back to the accumulated precipitation 
in the lower Mississippi and Ohio Valleys during January, February, and March. 

The flood of March and April. 1913, is in a class by itself; it responds to the condition; 
unprecedented rains plus high run-off with rivers at a medium stage. As in the case of mid- 
winter floods all of the factors in the expression for the 1913 flood are variables, and we 
therefore do not know whether or not the upper flood limit has yet been reached. 

Ordinarily the intensity of rainfall varies inversely as the duration; it is the exception 
that heavy rains continue over areas as large, for example, as the State of Ohio, for more than 
24 hours. In October, 1910. however, rains fell almost continuously over the lower Ohio 
Valley for a period of 4 days, although the period of great intensity realty covered only 3 days. 
As (he aggregate rainfall for this storm in some localities was equal to that which fell over 
parts of Ohio during March, 1913, a chart showing its horizontal distribution has been prepared 
and is presented as Chart No. 3 in the series of precipitation charts accompanying this report. 



ARE FLOODS IN THE OHIO INCREASING. 33 

We distinguish, of course, between the conditions of the soil as to moisture content and the 
stages of the streams in March and October, respectively. The bulk of the rain-in the October 
storm fell in the lower stretches of the Ohio Valley and doubtless the gage readings at Cairo, 111., 
and Evansville. Tnd.. will give us an idea of the amount of water which ran into the streams as 
a result of the heavy rains. The Ohio at Cairo, 111., on October 4. when the rains began, was 
at a stage of 12.2 feet; it rose steadily until the 11th, when a stage of 20. S feet was reached, 
after which it fell steadily until the end of the month. At Evansville. Ind., the Ohio rose from 
4.7 feet at 7 a. m. on October 4, to 26 feet at noon of the 9th. a total rise of 21.3 feet, due to the 
heavy rains of October 3-6. 

Considering that the rains on the dates mentioned were almost as heavy and continuous as 
those of March 23-27, 1913, it is interesting to note that the effect on the streams was scarcely 
noticeable as compared with the effect of the last-named floods. The maximum discharge in 
the Ohio at Evansville, in connection with the October, 1910. rains, was approximately 250,000 
second-feet on October 9, whereas, the approximate discharge at Evansville on March 25, 1913, 
the second day after the beginning of the rains, was, according to figures supplied by the 
United States Geological Survey, a little greater than the above figures, and reached a maximum 
value of 676,000 second-feet on April 5. From these figures it would seem reasonable to infer 
that in those seasons when the supply of water in both the soil and the streams is at a low ebb, 
as must be the case late in summer and throughout the autumn, little is to be feared from floods 
in the great rivers, due to heavy rains. Danger from local floods in the smaller streams is. 
however, an ever present possibility. 

Before closing the question of summer floods in the Ohio we would remark that practically 
the only semblance to a midsummer flood, as determined by the stages at Cincinnati. Ohio, 
occurred in August, 1875, when the river rose above the 50-foot stage and remained above that 
stage for five days, culminating at 55.3 feet on August 6. This flood was caused by heavy rains 
in July, which produced moderately high stages in the Ohio during the latter part of that 
month. A short period of heavy rains in the first part of August made the attainment of flood 
stages possible. In the smaller streams, as before indicated, flood stages are possible in the 
summer and autumn months, though relatively infrequent. 

Are floods in the Ohio increasing? That question has received a great deal of attention 
in recent years and several writers have answered it in the affirmative, citing the Ohio River 
as an example of greater flood frequency now than formerly. 

The writer has not been able to discover valid reasons for sharing in that belief; on the 
contrary a careful consideration of the data for the Ohio River leads to the opinion that while 
the data on the subject are inadequate to the formation of a definite conclusion, such as are 
available, can be so arranged as to point to either an affirmative of a negative conclusion, as 
will be shown later in this paper. 

It must be admitted that atmospheric precipitation is the sole source of the water which 
fills the streams. Meteorologists have known for many years that that element is variable 
both in time and space. The one other cause which enters largely in flood formation is the 
character of the surface covering in a watershed. It is conceivable that if the character of the 
surface cover be suddenly changed there might be a profound change in the run-off, but since 
all artificial changes in a river system are brought about slowly, and since some of them may 
augment the run-off while others may retard it, it becomes a matter of great difficulty to inte- 
grate the total effect of artificial changes in the watershed or stream flow for any given epoch. 
The Department of Agriculture is now carrying on experimental work in the Rio Grande 
Forest Reserve that, when completed will throw considerable light upon the subject. 

The writer has had occasion to examine the precipitation records of the United States 
as collected in the beginning by the Smithsonian Institution, and later by the United States 
Signal Service and its successor, the present Weather "Bureau of the Department of Agricul- 
ture. A brief reference to the results of that examination will be useful in connection with 
the present subject. 

14284°— 13 3 



34 THE FLOODS OF 1913. 

The total yearly precipitation varies greatly even at points but a short distance apart; 
it therefore becomes a difficult matter to express the monthly or annual abnormality for an 
urea so large as the United States. For convenience in discussion the latter area has been 
divided into 19 smaller climatic units and it is only through a combination of the results for 
the several smaller units that one can get an idea of the abnormality of the country as a whole. 
Since systematic observations began there has not been a single year when precipitation was 
above normal in all parts of the country, and but one year when it was deficient in all parts of 
the country in one and the same year. The abnormalities for the smaller climatic units, since 
the units are of unequal size, do not readily lend themselves to a combination, but considered 
separately they indicate the interesting fact that in the Rocky Mountain region from Montana 
southward to the Mexican border the number of years with precipitation above the normal are 
very nearly equal to the number of years with precipitation below the normal. East of the 
Mississippi, and in the Gulf and South Atlantic States, especially, the number of years with 
diminished precipitation are in a very decided majority. Further, it would seem that years 
of fat or lean precipitation do not follow any recognizable sequence and that the most probable 
value for any year is not the arithmetical mean or normal for the latitude, but a value near 
and slightly below the mean. On the other hand years of excessive rainfall do not occur with 
the same frequency as years of light rainfall ; heavy rains seem to be due to extraordinary and 
probably world-wide temperature and pressure relations which appear to be the cause of un- 
usual storm movement with its attendant precipitation. Thus in 1912 the prevalence of an 
unusually large number of southwest storms and the rains attending them produced the great 
flood in the lower Mississippi Valley of 1912. 

In a previous discussion of the subject of secular variation of precipitation 1 the writer 
prepared from the records of three stations in the middle Ohio Valley, Cincinnati, Portsmouth, 
and Marietta. Ohio, a diagram which he called " progressive averages of precipitation for the 
Ohio Valley." The curve in that diagram has been brought down to date and is presented below 
in Diagram II. It was prepared as follows: The annual precipitation at the three points 
named was combined in a single mean, which may be considered as a regional mean for that part 
of the Ohio Basin between Marietta and Cincinnati. It probably represents with considerable 
accuracy a region of 300 or 400 miles in all directions from the river as a central point, but it is 
not claimed that the figures represent the precipitation of the entire watershed. 

The heavy horizontal line in the diagram represents the normal precipitation ; where the 
curved line passes above the horizontal line precipitation was above the normal by amounts 
shown in the scale on the left margin. The diagram is conveniently read by considering the 
peaks as wet spells and the valleys as dry spells. This diagram is similar to others of a like 
character which have been constructed for different sections of the country in the fact that it 
shows that years of fat and lean precipitation follow each other at very irregular intervals and 
further that the period of fat or lean rains may extend over very unequal periods; thus the 
period 1841-1849 in the diagram was a period of fat rains which has not since been equaled. 
Likewise the memorable drought of 1838 in the Ohio Valley is indicated by the character of the 
curve for 1837-1839. Another important point to remember is that the peaks of fat rains in the 
Ohio Valley do not ordinarily coincide in point of time with similar peaks for the region Avest 
of the Mississippi or along the Gulf of Mexico. In other words, the variation of precipitation 
in regard to space is largely local rather than general. The valuations with respect to time 
are exceedingly irregular. At this point it is proper to mention that the calendar year is much 
too great a unit for comparative purposes; unfortunately the labor of assembling the data in 
a more convenient unit is so great as to be prohibitive. 

In Table No. 1 we have assembled all of the floods of consequence in the Ohio River from its 
source at Pittsburgh, Pa., to its mouth at Cairo, 111. We will now proceed to consider that 
record in connection with the rainfall curve of Diagram II. 

1 Bulletin 1), Weather Bureau. 1897, i>. IS. 



ABE FLOODS IN THE OHIO INCREASING. 



35 



In the endeavor by others to show that floods in the Ohio are increasing the record of 
river stages of 22 feet or more at one point in the river, Pittsburgh, Pa., for the 40 years 
1870-1910 has been considered. It was assumed that a stage of 22 feet, which, by the way, is 
the point at which the river begins to overflow its banks and is known as the flood stage, con- 
stituted a flood. The number of these so-called floods which occurred in the first half of the 
40-year period was compared with the number which occurred in the second half. The result 
of the count was that the number in the second half of the period considerably exceeded the 
number in the first half; therefore the conclusion that floods in the Ohio River are increasing. 



to 

UJ 

+ 7 
+ 6 
+ & 

+3 
+Z 
+1 

-/ 
-2 
-3 
-¥ 
-S 
-6 
-7 

+ 7 
+ 6 

+ + 
+3 

+ 2 

+ / 



-/ 

-2 

-3 
-V 
-6 
-6 


k 
2 
S 


<0 
"3 




s, 


\ 

> 


§ 


5 


$ 


b 
^ 


vo 


N 

V 


<0 




1 


^ 


£" 
^ 


5 


X 


X 


« 


K 

^ 




$ 


<5> 
vo 




v<? 


vo 


vo 


»0 

Vfl 


V© 
Vo 


vo 


CO 

vo 


vo 




fc 


a 


K 


* 








































































































" 
















































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































fc 




IS 




55 




Co 


2 


i 


»0 




N 

«0 


«0 
CD 


«0 


<0 










<0 






«0 


CS 

c> 






Ci 


2 




>0 

O 


vo 

V^ 




oo 

Ci 






^ 


S 






























































































































































































































































































































































































































































































































































































































^N 


v, 




































































\ 




/ 








































































\ 


/ 


r 































































































































































































































































































































































































Diagram II. — Progressive averages of precipitation in the Ohio Valley. 

There are certain obvious objections to this procedure ; in the first place, a rise in the river 
to the flood stage does not constitute a flood; the flood stage is simply the point on the gage 
when the river begins to overflow and cause damage to property on its banks; in the second 
place, as has already been pointed out in this paper, a river in flood at one single point in its 
course is not necessarily in flood at other points; therefore it is not proper to determine 



36 



THE FLOODS OF 1913. 



flood frequency from a single point in a river's course; in the third place, no account has been 
taken of the precipitation during the period under consideration. 

By reference to Diagram II we see at once that there was a period of dry years extend- 
ing from about 1869 to 1880, with the single exception of two years of about normal rainfall 
in 1875 and 1876. The deficiency for 1870 and 1871 was well marked and extended over the 
entire watershed. We are therefore justified in the assertion that the lack of floods in the Ohio 
above Cincinnati during the 10 years 1870-1880 was due simply to a term of dry years. Con- 
sidering further Diagram II, we see that a period of increased precipitation set in in 1880, 
culminating in 1882. The well-known floods of the early eighties in the Ohio Valley have 
already been described in this paper. The next period of heavy precipitation centered in 1890, 
but it was confined practically to a single year and caused serious floods only in the lower 
stretches of the river. (See Table No. 1.) 

The precipitation after 1890 Avas markedly deficient, especially during the drought years 
of 1894-95. A second short period of abundant rains centered in 1897-98. Serious floods 
occurred in both of these years. Then followed a term of dry years, which continued until 
1906. No serious floods occurred until 1907. The agreement between the record of precipita- 
tion and flood frequency is seen to be as close as might be reasonably expected. Let us next 
consider the record of serious floods in the Ohio as contained in Table No. 1. 

Dividing the 40-year period into two equal portions, it is seen, as before stated, that the 
greater number of floods occurred in the second half; but we have just shown that the absence 
of floods between 1869 and 1880 was due to a lack of precipitation. If we now disregard the 
first 10 years of the period and divide the remaining 30 years into two equal periods of 15 years 
each and count the floods therein, we have the result shown in tabular form below: 

Number of severe floods in the Ohio River during the 40 years, 1870-19J0, in two periods of 20 years each and 



nvi/vi uit i my i lit' J/u ycwis, ±oiv—±vju, 

also in two periods of 15 years each. 



Stations. 



Pittsburgh 
Cincinnati. 
Louisville. 
Evansville 



First period 
of 20 years, 
1870-1890. 



Second 
period of 
20 years, 
1891-1910. 



First period 
of 15 years, 
1881-1895. 



Second 
period of 
15 years, 
1896-1910. 



The Cairo stages are not included in this table since they are influenced by causes other than 
the amount of water coming down the Ohio and therefore would be misleading. 

The above is a very interesting and instructive table; we see at once that if we use the 
Pittsburgh stages as indicative of the number of floods in the Ohio there is a preponderance of 
floods in the second 20 and 15 year periods, respectively, but if we use Cincinnati the 
division into 20-year periods shows about an equal number in each; if we disregard the first 
10 years of the period and divide the remaining 30 years into two equal portions, we find the 
result of the Pittsburgh count is reversed, and that the preponderance of severe floods falls 
within the first period. The same results are reached if we consider the Louisville record, but 
Evansville, the next station below Louisville, coincides with Pittsburgh, but for entirely 
different reasons. 

The facts here given and the interpretation put upon them will have served the purpose 
of this paper if attention be focused upon what seems to be an obvious conclusion, viz, that flood 
frequency is primarily due to the distribution of precipitation as regards both time and space, 
and that there is urgent need of accurate measurements both of precipitation and stream flow 
for the next 50 years or longer before conclusions the one way or the other may be reached. 



PKOBABLE MAXIMUM FLOOD IN THE OHIO AT PITTSBUEGH. 37 

PROBABLE MAXIMUM FLOOD IN THE OHIO AT PITTSBURGH. 

In approaching a problem of this character we can 'be guided only by observation and 
experience. The Ohio at Pittsburgh has been under systematic daily observation a little more 
than 60 years, while records of floods dating back to the beginning of the nineteenth century 
are available. The extreme flood height in the last 100 years was 35.5 feet, on March 15, 1907. 
That stage was produced not by torrential rains over the entire watershed, but by the concur- 
rence of two favorable conditions, the first of which was effective over a comparatively small 
portion' of the entire Allegheny-Monongahela Basin, viz, the watersheds of thei Kiskiminetas 
and Youghiogheny Rivers. These two conditions were, first, the occurrence of a moderately 
heavy fall of snow on March 10, 1907, over the watersheds above named; second, the occurrence 
two days later of rain with rising temperature. The rain, which began on the 12th, was prac- 
tically continuous for 48 hours. 2.41 inches falling at Pittsburgh, Pa., and the average for the 
entire watershed was 1.97 inches. While the rainfall extended over the entire watershed, ii was 
heavy only in the neighborhood of Pittsburgh. The snow that made extremely high water 
probable was likewise confined to the watersheds of the Kiskiminetas and Youghiogheny. 
While the synchronism in the time of the several flood waves reaching Pittsburgh in this case 
was not perfect, it was nearly so. and there was produced the highest water of a century on the 
Pittsburgh gage on barely 2 inches of rainfall on the average for the entire watershed. 

The midwinter flood of 1884 was likewise due to a winter thaw in connection with only 
moderately heavy rains, but the volume of water which passed down the Ohio River in that 
flood was greater than in the 1913 flood, although the rainfall over the basin was probably not 
more than half as great as in 1913. The record of the 1884 flood illustrates the potentiality of 
snow covering as a factor in flood causation, and this fact is again emphasized in the record of 
the flood of March 1, 1902, when a stage of 32.4 feet was reached at Pittsburgh, Pa., on an 
average rainfall over the entire watershed of three-quarters of an inch. In this case the ice 
came out of the rivers converging at Pittsburgh separately, else a still higher stage might have 
been recorded. 

The danger of stages above 36 feet on the Pittsburgh gage seems to lie mostly in the 
probability of ice gorges in both rivers breaking up simultaneously. That long-continued rains, 
such as fell over Ohio and Indiana in March, 1913, may occur over some portion of the Alle- 
gheny-Monongahela watershed within the next hundred years is within the range of probability, 
but that such rains will be uniformly heavy over the entire watershed We consider only as a 
remote possibility. We consider the formation of ice gorges and their simultaneous breaking 
up in connection Avith spring rains to be the greatest flood menace in the upper Ohio Valley. 

Perfect synchronism in the breaking up of ice in the streams above Pittsburgh would 
doubtless cause a stage close to 38 feet on the Pittsburgh gage, and we would consider that figure 
as the probable maximum flood stage at Pittsburgh. However, if the watersheds of all the 
rivers above Pittsburgh were to receive as great a rainfall as occurred over Ohio and portions 
of Indiana in the flood of March, 1913, it is probable that a stage above 40 feet would be reached 
at Pittsburgh. 



FLOOD IN THE LOWER MISSISSIPPI RIVER. 



The condition of the Mississippi River as to gage heights on March 31, 1913, or when the 
crest of the Ohio flood was eight to ten days distant, is shown in the small table below; : 



Station. 



St. Paul, Minn . . 
Dubuque, Iowa. 

St. Louis, Mo 

New Madrid, Mo. 
Memphis, Term. 
Vicksburg, Miss. 
New Orleans, La 



Flood stage. 


Stage Mar. 
31. 


Feet. 


Feet. 


14 


1.4 


18 


10.2 


30 


23.2 


34 


39.7 


35 


36. 


45 


39.6 


18 


14.5 



Difference. 

Feet. 
-12.6 

- 7.8 

- 6.8 
+ 5.7 
+ 1.0 

- 5.4 

- 3.5 



Thus it will be seen that the river was considerably below the flood stage at St. Louis and 
thence northward to St. Paul, but it was above flood stage at Memphis and thence to Cairo, 
and rising slowly from Memphis to New Orleans. Moreover, the western tributaries were at 
moderate stages, and neither high water nor floods were impending in any one of them. It 
was apparent at the outset, therefore, that, barring additional precipitation, the only water to 
consider in estimating the volume of the flood in the lower Mississippi was that coming out 
of the Ohio. As soon as the probable output of the Ohio could be determined it became evident 
that the flood of 1912 in the lower Mississippi would be equaled, if not overtopped, and warnings 
to that effect were issued by the Weather Bureau official at Memphis as early as March 27, 
1913. about two weeks before the arrival of the flood crest. Warnings were issued at the New 
Orleans station on March 31 and repeated at intervals as the changing conditions demanded 
until May 14. The river at Memphis reached the flood stage of 35 feet on March 30 and 
continued to rise about a foot per day until April 10, when a maximum stage of 46.5 feet, the 
highest of record, was attained. This stage was a foot and two-tenths above the high water 
of 1912. On April 10 a break occurred in the levee at Wilson, Ark., above Memphis. The 
water escaping through this break, in connection with the flow through a previous break at 
Graves Bayou, on the Arkansas side below Memphis, caused a fall in the river at Memphis of 
a foot between 8 a. m. and 4 p. m. of April 10. There was no recovery in the river from the 
effect of the above-mentioned breaks, and the river passed below the flood stage at Memphis 
by April 29. 

The situation at Memphis just before the break occurred was indeed ominous. The levees 
had withstood a flood earlier in the year and were for the second time within a space of less 
than two months again subjected to the strain of a volume of water greater than ever before 
experienced ; portions of the city were already flooded ; gas was shut off ; westbound train 
serA'ice was discontinued; when to add to the peril of the situation heavy rains began to fall 
in Arkansas, portions of Louisiana, Mississippi, and in western Tennessee. At Little Rock 9.56 
inches fell in 24 hours ; curiously enough the effect of that great amount of rain on the Arkansas 
at Little Rock was not great, the total rise amounting to but 9 feet five days later. Evidently 
the downpour was local in character. At Memphis there was a fall of 3.22 inches in 24 hours 
on April 9, the day previous to the crest stage. 

The crest stage at Memphis, had not the levees given way, would probably have been 

somewhat higher than was actually recorded. (On this point see p. 90.) 

39 



40 



THE FLOODS OF 1913. 



Comparative statistics of floods in the lower Mississippi River published in Weather Bureau 
Bulletin Y show that the crest stages of the 1913 flood exceeded previous crest stages from New 
Madrid. Mo., to Helena. Ark.; also at Natchez, Miss. 

The facts which conspired to produce lower crest stages in the New Orleans district than 
were experienced in 1912 were (1) a very slow return of Lake St. John crevasse water, thus 
enabling the bulk of the flood waters in the Mississippi to pass the mouth of the Red River 
before the arrival of the crest of the crevasse water, and (2) a short period of northerly winds 
about May 11, 1913, which tended to augment the discharge of the flood waters into the Gulf 
of Mexico. 

The duration of the flood is shown in the small table below, in which similar statistics for 
1912 are given for comparative purposes: 

Duration of 1.912 anil WIS floods. 



Stations. 



Number of days above flood stage. 



1912 



1913 



First, 
flood. 



Second 
flood. 



Total. 



Cairo, 111 

Memphis, Term. 1 ... 

Helena, Ark 

Arkansas City, Ark 

Vicksbnrg, Miss 

Natchez, Miss 

Baton Rouge, La. . 
Donaldsonville, La 
New Orleans, La.. 



47 
55 
59 
54 
63 
57 
59 
53 
46 



1 On Mar. 1, 1912, flood stage changed from 33 to 35 feet; the latter value was used in this table. 

In forming the table the two floods in 1913 are considered separately and the total number 
of days above the flood stage for both floods is given. Thus we discover that so far as dura- 
tion is concerned the combined number of flood clays in 1913 is somewhat short of the number 
of flood days in 1912. It seems, therefore, that notwithstanding the higher crest stages in 
1913 than in 1912 the volume of the latter flood was really the greater of the two. 

The overflowed area in 1912 also appears to have been considerably greater than in both 
floods of 1913. 

Area overflowed. — The area of overflowed lands, whether from crevasse water or backwater, 
was less from the two floods of 1913 than from the single flood of 1912. 

In the Memphis district, Cairo to Helena, the levees did not break in January. 1913, and 
consequently very little damage from overflow was sustained. The only crevasse of any im- 
portance in the January, 1913, flood occurred in the Vicksbnrg district near Beulah, Miss., the 
scene of a serious break in the levee in 1912. Repairs to the 1912 break had not been wholly 
made when the January, 1913, flood reached that point. At first the break had a width of 
but 200 feet, but a rise of 6 feet additional in the river caused the break to widen to about 
1,000 feet in spite of efforts to restrain it within its original bounds. Efforts to prevent the 
break attaining a greater width were, however, successful, and the resulting overflow was only 
about half of that which occurred in 1912. While the flood waters were still pouring through 
the crevasse a successful attempt was made to close it. The different stages in the attempt 
are well shown in the series of halftones reproduced in this report. 

Figure 4 shows the crevasse as it appeared in February. 1913. 

Figure 5 shows the trestle constructed of piling, and the rock dumped into the break. 

Figures 6, 7, and 8 show, respectively, the final steps of clumping earth on both sides of 
the rock fill. 




o t 



10 »> 



Chart 5, Part 2. 



Flooded district (in red) 




MONEY LOSS DUE TO FLOOD. 41 

The closing of the .Beulah crevasse prevented the overflow of about 1,000 square miles 
that was inundated in 1912. 

The overflowed areas in the second 1913 flood comprise portions of southeastern Missouri, 
especially the counties of Mississippi. New Madrid, Dunklin, and Pemiscot. The total area 
ovei'flowed in Missouri and Arkansas as computed by the Mississippi River Commission being 
2,454 square mines. In Arkansas the overflowed counties were Mississippi, parts of Poinsett, 
Crittenden, St. Francis, Lee, and Desha. The latter was overflowed in part from a break in 
White River levee. 

In general the overflowed region in Arkansas in 1913 was not greatly different from that 
of 1912. (For details see report of Section Director H. F. Alciatore, p. 97.) 

The overflowed areas in Louisiana were not so large in 1913 as in the previous year. In 
the parishes of northeastern Louisiana, viz, West Carroll, East Carroll, Morehouse, and Rich- 
land there was considerable overflow, principally along the bayous in those districts. In the 
central-eastern parish of Concordia the overflow was general and it extended westward into 
Catahoula Parish almost to the lake of that name. There was also a third region of overflow, 
viz. along the Atchafalaya River below Melville in the parishes of St. Landry, St. Martin, 
Iberville, Iberia, and Assumption. 

On the left bank of the Mississippi River there were overflows in Hickman and Fulton 
Counties in Kentucky; also in the Reelfoot Lake region of Tennessee and in small portions 
of Tennessee counties bordering the river. 

In Mississippi the counties which were overflowed in whole or in part were, from north 
to south, Bolivar, Sunflower, Washington, Sharkey, Yazoo, Issaquena, and Warren. There 
was also some overflow along the Yazoo River in the counties of Tallahatchie, Leflore, and 
Carroll. 

A small scale map showing overflowed areas is presented as Chart No. 5. This map has 
been reduced in great part from larger scale maps prepared by the Mississippi River Com- 
mission, to whom we are indebted for most of the information. 

MONEY LOSS DUE TO FLOODS. 

The elements of fixed capital that suffered in the floods of March and April, 1913, are 
chiefly those of railway lines, telegraph and telephone systems, bridges and highways, dwell- 
ings and other buildings such as power plants, and factories. 

Practically all of the above-mentioned elements suffered ioss that will require the outlay 
of additional capital either to replace the tangible property lost or destroyed, or, in case the 
loss was not total, to restore the system to its original standard of efficiency. In any event 
the original investment has been increased by just so much as it cost to replace the property 
destroyed. Loss was also sustained due to interruption of trade or traffic, the failure of pros- 
pective crops, and the suspension of business; moreover, vast sums of money and a great deal 
of labor was devoted to strengthening the levees along the great rivers and in other protective 
measures, none of which appears in the final summaries of loss. 

Much labor and considerable time has been devoted to the collecting of the statistics of 
loss here given ; the most of which are as accurate as could be had except by an actual census 
of the situation. 

In the case of loss to railroads the majority of the facts were supplied by the responsible 
authorities directly concerned; in a few cases, however, in the absence of definite figures, re- 
course has been had to a carefully prepared estimate of the loss. There is therefore an element 
of uncertainty in some of the amounts that may amount to at least 10 per cent; in the state- 
ments of loss of crops both matured and prospective ; and of loss due to suspension of business 
the uncertainty, from the nature of the respective cases, is much greater. 

We present first a statement of loss suffered by railroads and telegraph and telephone 
systems arranged by States wherever possible. In the Mississippi Valley States of Missouri, 
Arkansas, Louisiana, Mississippi, and Tennessee, it has not been possible to distinguish as 
between States ; the total loss is therefore grouped under the general head Mississippi Valley. 



42 



THE FLOODS OF 1913. 
Loss to railroads, telegraph and telephone companies. 



State or district. 


Loss to rail- 
roads. 


Loss to tele- 
phone and 
telegraph com- 
panies. 


Ohio 


$6, 493, 555. 00 
4,812,805.00 
1,391,544.00 
150, 000. 00 
3,120,661.00 
' 200, 000. 00 


$1,982,756.00 
9 000.00 


Indiana . , , 


Illinois 




Kentucky 


5,000.00 
6, 423. 00 


Mississippi Valley 


New England 








Total 


16, 168, 565. 00 


2, 003, 179. 00 





1 Estimated. 



Losses other than to railroads, telegraph and telephone companies. 



City or district. 



Ohio River districts: 

Pittsburgh district 

Parkersburg district 

Cincinnati district 

Louisville district 

E vansville district 

Cairo district 

Mississippi River districts: 

Memphis district 

Vicksburg district 

New Orleans district 

Cumberland River at Nashville 

Upper Hudson and Mohawk Rivers. Albany district 

Connecticut River Valley and Vermont 

White River, Ind 

Wabash River, Ind 

Smaller rivers of Ohio 

Total 



Tangible 
property, 
buildings, 
bridges, high- 
ways, etc. 



$2, 



500. 000 
350, 000 
530, 200 
800, 000 
700, 000 
S10, 000 

203, 000 
500, 000 

13, 750 
157, 500 
950, 000 

37, 500 
659, 855 
000, 000 
865, 526 



122,077,331 



Matured 
crops. 



$125,000 

214,040 
150, 000 
240, 000 
23, 200 



Farms and 
farm prop- 
erty, includ- 
ing prospec- 
tive crops. 



$25,000 



752, 240 



1, 125, 000 
150,000 

470, 000 
500, 000 
225,000 



262, 500 



1,412,800 



4,170,300 



Stock and 
movable 
property. 



$1,000 



205,000 

998,000 

125, 000 

12,000 

3,500 



51.150 



i 5, 002, 000 



6, 397, 650 



Suspension 
of business. 



$200, 000 

100, 000 

1, 360, 850 

500, 000 

500,000 



1,720,000 

350,000 

75,000 

23,000 

150,000 



622,600 
6, 394, 078 



11,995,528 



Total. 



$2, 725, 000 
2,451,000 
3, 891, 050 
1,300,000 
2,325,000 
1,290,000 

5, 605, 040 

1,625,000 

565, 750 

207, 200 

1,100,000 

37,500 

5, 596, 105 

10,000,000 

106,674,404 



145,393,049 



1 Distributed as follows: Farm buildings and fences, $1,352,000: soil washing, $3,300,000; driftwood and debris, S350.000. 

Grand total of loss from all causes, $163,564,793. 

The Work of the Weather Bureau in the forecasting of floods is frequently referred to in 
the reports of local forecasters and others in the section which forms the concluding part of 
this report. No effort has been made to reduce to dollars and cents the saving of property 
that resulted from the forewarnings of the floods herein described. That such forewarnings 
are valuable no one will deny, yet there is evidence to show that the full measure of benefits 
that should have been enjoyed from them was sacrificed, in some cases, because the personal 
opinion of the recipients differed from that expressed in the bureau's warnings. 

If it could be shown that not a single dollar is saved by the flood service of the Weather 
Bureau, the work would nevertheless be justifiable on the broader ground that it contributes 
to the general welfare of the people and conduces to the protection of human life. 



ACCOMPANYING PAPERS. 



DETAILED REPORTS OF FLOODS IN OHIO, INDIANA, ILLINOIS, 

KENTUCKY, NEW YORK, NEW ENGLAND, AND THE 

LOWER MISSISSIPPI VALLEY. 



43 



Chart 6, Part 2. Total precipitation in Ohio. 8 a. m., March 23, to 
8 a. m., March 25, 1913, inclusive (inches). 




Chart 7, Part 2. Total precipitation in Ohjo for 48 hours ending 
7 p. m., March 25, 1913 (inches). 




Chart 8, Part 2. Total precipitation in Ohio, 8 a. m., March 23, to 
8 a. m., March 27, 1913, inclusive (inches). 




Chart 9, Part 2. Total damage to highways and bridges in flood of 
March, 1913, as furnished by county commissioners in July. 1913 




., WILLIAMS i ^^^ . 
5000 J-2000O-1 
» L-] 

l.5j| DEFIANCE IHENRY 

5j_-9QQqO j 20000 

PAUL0INS 




vooo ! ""••J"* ! 55( ^' E ! LORA, 1 H^ 0( ^"i~''~~nTi.UMeu U .'j 
Hancock r 1 r - ] o° r l I I r— 1 



9700 f 

WYANDOT CRAWFORD 



??3W ! 



COLUMBIANA 



12700 



L CARROLL J ~" 1 

iNoneJ §| 



, 60000 i 45 000 Mf ! ^^ 

1 45264 V -1 "~ 

i HAI "' iN j 65000 [ 

8j mercer | gam j- -ri-A" ! ° Jmor _3?.5Q0_; 200000^-- 

| 7600 j -i- ,75000! "-ox LT «WMr ^ 

8! >j SHELBY o QKn - l" L H 15 0°00 i COSHOCTON j r 1 HARRISON j £ 

! i2oowr----S ON,ON ^";Z E i" ~i-2!5V- -=L^bbl±l 

■»« j j champaign ^3QQQ^ir««Lj LICK)Me euERNSEY I b£lmont 

Kr-^ifUr ' 3180 ° J M ~L 10 99P-j 5500 

■ clark' / « ^ <*£wwf 1 -1450000 

in / r !. 1550000 J h _ n tone! 

JTJ * f FAIRFIELD L LT^ «-*W!l'E^ MONROE 

FhIp,^ I 65000 H ; c ^L! 1 ^ffi? N l l_r-i 700 °, 




S! BUTLER i WARREN 

^Leooooo jL? 00000 
li 1M <~~y 



* 



|25flfej 



7T "" ""k 



977-.0 



15000! 



a i h tna 

v.Won | 30000 

. r 1 — 

I— i2C 

JACKSON j 

l^iaoa 



8000 






\. 



Total, $12,031,039. 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 



By J. Warren Smith, Professor of Meteorology. 



The flood of March. 1913. was due to an excessive rain falling upon a surface that was 
already thoroughly saturated. There was no snow upon the ground, and the surface soil was 
not frozen, so that the usual winter flood conditions were absent ; but the surface soil Avas very 
wet. and small streams and depressions were well filled by frequent precipitation during the 
first part of March. 

On Sunday, March 23, rain began to fall in northwestern Ohio at about 8 a. m., and in 
central districts during the middle of the forenoon. The rainfall during Sunday was about 2 
inches in the Maumee and Sandusky watersheds, in northwestern Ohio, and about 0.5 inch in 
central districts, but was very slight in southern counties. The rain ended by early evening 
in most of the State. 

Rain began to fall again soon after midnight and fell almost continuously through the 24th, 
25th, and 26th. By the morning of the 24th the total 24-hour rainfall was about 3 inches in 
the Sandusky, Maumee, and upper Great Miami River Valleys, but less than 1 inch in the 
southeastern portions of Ohio. 

During the day, Monday, March 24, the fall was heaviest in the upper Great Miami Valley 
and amounted to 3.7 inches at Piqua, Miami Count}^. 

The rainfall for the 24 hours ending the evening of the 24th averaged 1.9 inches in the 
Scioto Valley above Columbus, 1.9 inches in the Great Miami Valley above Dayton, 2.1 inches 
in the Little Miami Valley. 1.6 inches in the Sandusky Valley, and 1.3 inches in the Muskingum 
Valley above Zanesville. The greatest fall reported was 5.6 inches at Piqua. 

The total rainfall from Sunday morning up to the evening of Monday the 24th averaged 
3.6 inches in the Sandusky watershed. 3.1 inches in the Scioto above Columbus, 2.9 inches in 
the Great Miami above Dayton, 2.4 inches in the Little Miami, and 1.9 inches in the Muskingum 
above Zanesville. In Hancock and Wyandot Counties it was over 4 inches, and in Miami 
County 6.6 inches. 

The rainfall was very heavy during Monday night. At Bellefontaine, in Logan County, 
it was about 4.5 inches; in Richland County, 4 inches; Marion County, 3.6 inches; Shelby 
County, 3.5 inches ; and in Seneca County, 3.3 inches. 

The total rainfall up to the morning of the 25th is shown in chart No. 6. The precipitation 
on that chart is for 48 hours, and it was that rainfall which caused the flood waters in Indiana 
and western Ohio. Unfortunately not many of the observations are made in the morning, 
but the above-named chart shows a fall of 9 inches in Miami County, and 7 inches in Marion 
County, and 7.4 inches in Logan County. The fall was over 6 inches over a large portion of 
the north-central part of Ohio. 

The heaviest rainfall had moved eastward by the 25th, and the Muskingum Valley received 
more than the western watersheds. At Ashland, Ashland County, the fall from 12.30 p. m.. 
Monday, the 24th, until 12.30 p. m.. Tuesday, was 5.96 inches. The rainfall for the 24 hours 
ending the evening of the 25th was more than 4 inches over a large part of the upper Muskingum 
watershed, as well as in the central Great Miami Valley and the upper Scioto. 

45 



46 THE FLOODS OF 1913. 

The total rainfall up to the evening of the 25th is indicated in chart No. 7. The total 
rainfall from the morning of March 23 to the evening of March 27 is given in chart No. 8. 
It shows that the fall was over 10 inches over a large district. 

Table No. 2, page 64, gives the carefully determined average rainfall in each watershed 
for each 24 hours and for the 5 days. The area of the watershed is shown, as well as the 
calculated fall of water, in cubic feet per square mile. The Great Miami watershed (total) 
includes the White Water River in Indiana, where the rainfall was very heavy. 

The daily rainfall records will be found in Table No. 2, page 21. 

A very careful study of the run -off from the small valleys of the upper Great Miami River 
is being made by the Morgan Engineering Co., of Dayton, and their computations show that 
the rainfall over a large district east of Piqua and Troy must have been considerably greater 
than was recorded at any of the surrounding stations. 

OTHER HEAVY RAINFALLS. 

The heaviest monthly rainfall in Ohio since the records began in 1854 was in September. 
1866. At North Lewisburg and Urbana, Champaign County, the total fall for the month was 
15.'.) inches, while the average for 20 stations well distributed over the State was 9.7 inches. 

The following item relative to the high water of that month is from the twenty-eighth 
annual report of the Ohio Board of Public Works for 1866: 

Very extensive damage was done at Circleville on September 19. The water was 8 inches higher than 
in the great flood of 18G0, which was then said to be unprecedented in the Scioto River. The water flowed 
over the high canal bank for almost the entire distance between the lock at the west end of Circleville 
aqueduct and Foresman's mill, and near the last-named point the banks were entirely swept away. Several 
breaks occurred in the canal feeder between Columbus and 4-mile lock. Much damage was done to the 
canals in the Muskingum Valley. 

This rain gave the highest water at Dayton previous to March, 1913, but daily records are' 
not available. 

A careful tabulation has been made of all of the 24, 48, 72, and 96 hour rainfalls at all of 
the stations in Ohio from 1883 to date. This table is too long to publish in this report, but it 
shows that the only time when the rainfall approaches that for March, 1913, in area covered 
and intensity of fall was in October, 1910. (See Chart No. 3.) 

The rainfall for October 5-6, 1910, was 7.5 inches in Butler County; 7.7 inches in Warren 
County; over 6 inches in Montgomery, Clark, Champaign, and Delaware Counties; and over 
5 inches from Hamilton north to Shelby County and east to Richland County. The fall for 
October 4 to 7, inclusive, 1910, was not so heavy by a third as occurred on March 23 to 26, 1913, 
and the heaviest rainfall was in the lower portion of the Great Miami Valley instead of the 
upper watershed as in March, 1913. 

RISE OF THE RIVERS. 

The streams in the northwestern portion of Ohio began to rise rapidly during the night 
of the 23d, and by; the morning of the 24th the flood stage was passed at Fort Wayne, Inch, 
and at Upper Sandusky and Tiffin, Ohio. 

By noon of the 24th the Great Miami, Little Miami, and the Scioto Rivers had begun to 
rise at a rapid rate. These rivers continued to> rise during the night of the 24th and the 
morning of the 25th, and in the lower portions until the 26th. 

The Muskingum River did not begin to rise rapidly until the night of the 24th, and did 
not reach its highest stage until the 27th. 

The river gauge readings are shown in detail in Tables Nog. 3 and 4, and a further discussion 
of the rate of rise is given in the story regarding each watershed. 

From the figures of Table 2 the total amount of water falling on the several watersheds 
can be easily determined and the probable run-off calculated. In the Scioto watershed above 
Columbus the total fall of water from the 23d to 27th, inclusive, amounted to 235,813.912,500 
gallons, or 31,441,855,000 cubic feet. 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 47 

DAMAGE AM) Miss I',V Till: FLOOD. 

The following statement gives the best figures of the loss and damage that it has been 
possible to collect. In many cases the amounts given are only estimates, but they are conserva- 
tive ones, and the total loss was undoubtedly far greater than these compilations indicate. 
(See also Table 5.) 

But no tables can tell all the tale; indeed, no pen can recount the story of horror, of suffer- 
ing, of countless acts of self-sacrifice and heroism. There were many instances of the savings 
of n lifetime being wiped out in an hour, of priceless pictures, books, and heirlooms washed 
away or ruined by the water and mud. 

Neither do the tables tell of light plants being out of use and cities and villages in darkness 
for a week or more, of water supplies being cut off, sewage plants out of commission, or I aim 
lands completely destroyed. The whole story is indeed a sad one, and undoubtedly the " Big 
flood of 1913 " will be a reckoning point for many years to come. 

There were over 100 municipalities in Ohio affected by the flood. The total population of 
the cities and towns most affected by the flood was 1.388,000. The total number of lives lost 
Tn the flood as near as can be determined was 467. The approximate number of residences 
flooded was 40,037, and the approximate number of houses destroyed was 2,220. 

Total estimated money loss in Ohio. 

Damage to public highways and bridges (see Chart No. 9) $12,031,039 

Damage to buildings and personal property 78,072,387 

Damage to farm buildings and fences 1,352,000 

Damage to farms by soil washing 3,300,000 

Damage to farms by driftwood and debris 350.000 

Crops destroyed or damaged 1,412,800 

Loss of live stock 234,953 

Cost of cleaning and repairs to machinery, etc 3,702. 100 

Damage to railroads, physical plants 6,051,300 

Loss to railroads, enforced suspension of business 6,000,000 

Damage to interurban eleetri<- roads 220,872 

Loss to same by suspension of business 173. 965 

Damage to street and suburban lines 221, 383 

Loss to same by suspension of business 200,000 

Damage to physical plants of telephone lines 110.000 

Loss to same by enforced suspension of business 20,113 

Damage to lines of Western Union Telegraph Co 150,000 

Total 113, 662, 918 

MATJMEE RIVER. 

The Maumee watershed covers the northwestern corner of Ohio, northeastern Indiana, and 
a small portion of southern Michigan. It has a total area of 0,344 square miles, 4,702 of which 
are in Ohio. It empties into Maumee Bay at the western end of Lake Erie. The main tribu- 
taries from the north are the Tiffin and St. Josephs Rivers, and from the south the Blanchard 
and St. Marys Rivers. The rivers of the southern watershed flow through a district that is 
extremely flat. The northern watershed is more rolling. 

At Montpelier on the St. Josephs River the water was 7 feet higher than has been 
experienced for some years. 

At St. Marys, near the head of the St. Marys River, the water was from 2 to 3 feet higher 
than before known, but only a small per cent of the city was flooded. There was much uneasi- 
ness, however, because of danger from the Grand Reservoir, which is situated about 2 miles 
west. Many people left their homes in this city, as well as in Celina at the western end of the 
reservoir, and fled to higher ground. This reservoir has an area of 1 5,748 acres and is the largest 
artificial lake in Ohio. It drains only about 03 square miles, but filled very quickly with the 



48 THE FLOODS OF 1913. 

heavy rainfall, and there was considerable washing of the banks. The spillway from this 
reservoir is at the western end. and the surplus water flows into the Wabash River. 

At Wapakoneta on the Auglaize River the water was 3 feet higher than any previous 
record. The damage was not large. 

Findlay is on the Blanchard River, a branch of the Auglaize. The water here was 3 feet 
higher than during the previous highest flood, on April 1, 1904, and 60 per cent of the city was 
flooded. There was one death, and 3,000 people were driven out of their homes by the water. 

Ottawa is also on the Blanchard River and is in the same flat region. Fully 95 per cent of 
this city, which has a population of 2.200, was flooded, although in most of the city the current 
was not swift and little damage was done to buildings. 

Lima is on the Ottawa River in central Allen County. Where unobstructed the water was 
1.5 feet higher than the former highest record, on April 1, 1904. There was one death and 
considerable property damage in this city. 

Two hundred and sixty-eight houses were flooded at Defiance and 150 at Napoleon, both 
on the Maumee River.'and there Avas one death at Napoleon. 

The damage along the smaller streams that empty into Lake Erie was not generally large 
outside of the injury to railroads and bridges. The Portage River at Bowling Green was 3 
feet higher than any previous record. In Lorain County a railroad engine broke through a 
weakened trestle and the crew of three men was drowned. In Cleveland the lowlands alone: 
the Cuyahoga River were overflowed and a large amount of lumber and other light material 
was carried downstream and into the lake. The wooden wharves along the river banks and a 
large amount of freight in cars standing in the railroad yards were damaged. Damage was 
done to bridges by the breaking away of steamers from their moorings. 

A good deal of damage was done at Akron. Summit County, on the Little Cuyahoga River, 
by the breaking of the banks of the largest of a string of artificial lakes known as the Portage 
Lakes. It is called the East Reservoir, and the bank was cut where the feeder from the 
Tuscarawas River enters. The power equipment and storerooms of the Goodyear tire and 
rubber plant were flooded, causing a loss estimated at from $50,000 to $100,000; one of the city 
playgrounds was damaged to the extent of $12,000; bridges were carried away; and 15 to 30 
houses were demolished. 

At Akron, as well as at many other points along the watershed between Lake Erie and 
the Ohio River drainage areas, the unusual spectacle was presented of large damage from flood 
at the crest of the watershed. At Bellevue, at the junction of Sandusky and Huron Counties, 
the flood did not recede for nearly five weeks, the first floors of many of the houses being 
flooded for this period. This village, with a population of about 4,149, is not located on a 
stream, but the surface water is usually drained into natural " sinks " or else drilled holes. 
Because. of the excessive rainfall, the ground water was rapidly raised and the "sinks" which 
take the water in under ordinary conditions became springs. 

SANDUSKY RIVER. 

The Sandusky River rises in the western portion of Richland County, flows west through 
Crawford and into Wyandot County, then north through Seneca and Sandusky Counties, and 
empties into the west end of Sandusky Bay in southern Lake Erie. The river is 115 miles 
long and drains 1,554 square miles. Its headwaters are 709 feet above the mouth. The valley 
is small, having a depth of from 20 to 25 feet and an average width of one-fourth mile or less. 
There are quite a number of dams along the main stream, the largest being the Ballville Dam, 
2 miles above Fremont. 

The heavy rainfall caused a rapid rise of all the small streams in this valley, and on the 
morning of the 24th the river at Upper Sandusky was 3.8 feet above the flood stage and was just 
at flood stage at Tiffin. It rose steadily and rapidly at Upper Sandusky during the next 24 
hours and reached 19 feet at 9 a. m. March 25. This was 6 feet above the flood stage, and 3 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 191.3. 49 

feet higher than occurred February 3. 1883, and 4 feet higher than on March 4, 1847, the 
previous high-water records. Inasmuch as the river is some 25 feet or more below the general 
level, very little damage was done either in Wyandot or Crawford Counties. 

At Tiffin, Seneca County, the flood was much more serious. The river flows through the 
center of the city from southwest to northeast, and a fair-sized branch comes in from the 
southeast. The buildings encroach very decidedly upon the stream channel, and this is very 
narrow and crooked. It is crossed by six bridges, none of which was high enough or wide 
enough to prevent damming. 

The river was at flood stage (7 feet) at Tiffin at 7 a. m. March 24 and had risen to 7.8 feet 
at 1 p. m. At 7 a. m. the 25th the gage was 12.5 feet, and at 11 a. m. it was 14 feet, and the 
bridges and streets near them were impassable. At 1.40 p, m. the first bridge went out; and, as 
the current is very swift through the city, the other bridges and buildings nearest the channel 
were carried away in rapid succession. The height of the water at the river gage was 19.4 feet 
at 6 a. m. March 20, or 8 feet higher than ever before recorded. At the Hubach brewery it was 
28.9 feet, or 10.3 feet higher than the previous record, and at the waterworks 9 feet above the 
highest before known. 

The observer, Prof. T. H. Sonnedecker, has made some investigations and concludes that 
this was the greatest flood in that valley since it has been settled by the white man and probably 
the greatest in this geological era. 

At Fremont the river passed the flood stage, 10 feet, at about 9 a. m. of the 24th. At 
7 p. m. it was 12.4 feet ; at 7 a. m. of the 25th, 13.5 ; at 3 p. m., 17.5 ; and by noon of the 26th it 
had reached 21.5 feet, or 3.7 feet above the record of February, 1884. It remained practically 
stationary until the following morning. The river gage is on the downstream side of the 
bridge, and it was 1 foot higher above than below the bridge. 

About 35 per cent of the city was flooded, which includes a large proportion of the business 
district. The loss to merchandise stock and buildings was estimated at $262,000. The loss on 
the Ballville Dam was $85,000. Three men were drowned at Fremont. 

MAHONING EIVEE. 

Xumerous power and water-supply dams on the Mahoning River makes this river a 
series of pools and riffles from its source to the Pennsylvania line, greatly affecting the run-off 
of the stream. 

At "Warren, in Trumbull County, the river was 4.2 feet higher than before recorded. Two 
persons were drowned and 150 houses were flooded. The highest water was at 6 p. m. March 26. 

At Garrettsville, Portage County, the water was 1.5 feet higher than any previous record. 

Niles, Trumbull County, the river was 6.8 feet higher than in January, 1904, the previous 
high water. From 10.15 p. m., March 24, until 8.15 a. m., March 25, the water rose at an 
average rate of 6 inches per hour. 

At Youngstown, Mahoning County, the highest water was at 11 p. m. March 26, when it 
reached 22.9 feet, or 7.1 feet higher than the previous record, January 21, 1904. 

HOCKING RIVER. 

The rainfall was not particularly heavy in southeastern Ohio and the flood was not so 
serious in the Hocking Valley as elsewhere. At Logan, Hocking County, the river was not so 
high by 1.5 feet as was recorded in 1884. 

WHITE WATER RIVER, IND. 

The highest stages for this stream and its tributaries ever known were reported during the 
flood period March 23-27, 1913. The rainfall in this valley as reported by cooperative 
observers of the Weather Bureau for this period was as follows: Cambridge City, 9.38 inches; 
14284°— 13 1 



50 THE FLOODS OF 1913. 

Connersville, 9.98 inches: Richmond, 11.15 inches. On March -25. Cambridge City recorded 
5.70 inches: Connersville. 5.07 inches. On March 24, Richmond reported 5.30 inches and on 
the 25th 4.17 inches, making a total of 9.47 inches for the 48 hours. Xo snow was on the 
ground to augment the stream flow, but the ground was well soaked by previous rains when 
the heavy downpours occurred and the run-off was quite rapid. Perhaps it was equal to what 
it would have been had the ground been frozen. 

The county commissioners of Union, Henry, Fayette, Wayne, and Randolph Counties 
report a total loss of $101,100 in damage to county and municipal bridges and a loss of $67,500 
in damage to public highways. This makes a total loss of $228,600 in damage to public high- 
ways and bridges. 

Interurban electric lines operating in this valley report a loss of approximately $35,000 
to their lines, bridges, and rolling stock and about $20,000 loss in business due to enforced 
suspension. 

The correspondent at Cambridge City reports the loss of two lives due to the flood. The 
Red Cross Society reports 15 deaths at Brookville, Franklin County. 

THE GREAT MIAMI RIVER. 

The drainage basin of the Great Miami River lies in southwestern Ohio and southeastern 
Indiana and has an area of approximately 5,400 square miles. One-third of the total area is 
in Indiana. The river is formed in Logan County by two small streams rising in Auglaize 
and Hardin Counties. It flows in a southwesterly direction and joins the Ohio River at the 
Indiana State line. The length of the river is 140 miles. The main tributaries are the Still- 
water River from the west and the Mad River from the east, both entering the Great Miami 
just above Dayton. 

The valleys above Dayton are narrow and comparatively shallow, while below Dayton the 
valley is broad and open. The average fall of the river is 4.1 feet per mile and is steepest in 
the upper part, especially in the Mad River drainage area. 

The Lewiston Reservoir is in Logan County and is often designated as the source of the 
Great Miami. It has an area of about 6.134 acres and drains a watershed of 100 square miles. 

The Loramie Reservoir is situated in the western portion of Shelby County and feeds into 
the Great Miami between Sidney and Piqua. It has an area of 1,900 acres and drains 70 
square miles. 

Both of these reservoirs were soon filled at the time of the heavy rainfall and there was 
grave danger that the banks would give way. There was no serious break in either one, how- 
ever, and on the other hand they did very little to minimize the flood conditions. 

Sidney is the first large town on the upper Miami. It is located in Shelby County and 
has a population of nearly 7,000. Only about 6 per cent of the city was flooded by the high 
water, and no lives were lost. But three residences Avere destroyed, although 230 were in the 
flooded district. The water was the highest at about 9 a. m. March 25, when it was 4 to 6 feet 
higher than before recorded. The damage to buildings and personal property is given as $35,000. 

The damage was much more at Piqua than at Sidney. Fully 50 per cent of Piqua was 
flooded, and 44 lives were lost. The number of residences flooded was 2.000, and 100 or more 
were destroyed. The river gage at Piqua was at 8.7 feet on the morning of the 24th, and the 
water had risen only to 11 feet at 5 p. m. It rose more rapidly after that hour and reached 
the flood stage of 14 feet at 9 p. m. The rise from that hour until 10 a. m. on the 25th Avas 
10 feet. The river was about stationary at 24 feet until 2 p.m.; then fell steadily. By the 
27th it had fallen to the flood stage. About 2,000 people were rendered homeless at Piqua. 
During the night of the 24th the water broke through the leA^ee that protected the portion of 
the city east of the Miami and Erie Canal and also flooded East Piqua. Later the water came 
clown through the main portion of the city. The water was 6 feet deep in some of the business 
districts, and the flood damage was very great. 




iZ S 



>,o 




FIG. 10.— STEELE HIGH SCHOOL BUILDING, DAYTON, OHIO. 
Showing destruction due to undermining foundation by water. 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 



51 



Troy is the next town of moderate size below Piqua. It has a population of over 6,000 
and covers about 2 square miles. Fulty 75 per cent of the city was flooded. Water affected 
nearly 1,000 houses, and 50 were washed away. The highest water was about noon March 25. 
Sixteen lives were lost. The loss at Troy was, to business houses. $35.000 ; factories, $150,000 ; 
residences and personal property, $175,000. 

Springfield is a city of nearly 50,000 population and is situated on the Mad River in 
central Clark County, 25 miles above Dayton. A moderate amount of damage was done in 
the city, as most of it is well above flood waters, and only one life was lost. 

Dayton is in Montgomery County, 33 miles below Piqua and 77 miles above the Ohio River. 
The drainage area above the city is 2,558 square miles. The river is 600 feet wide at average low 
water. The central portion of the city is located on the flood plain of the river, which at this 
point is about 3-J miles wide. Approximately 50 per cent of the city is built on this flood 
plain. The following report is made by Mr. H. F. Alps, local forecaster at Dayton : 

REPORT OF FLOOD AT DAYTON, OHIO, MARCH 25, 1913. 

The most destructive flood in the history of Dayton occurred on Tuesday morning, March 25, 1913, fol- 
lowing about 36 hours of unprecedented rainfall over the Great Miami River watershed. The center of the city 
was submerged to a depth of 9 feet, and in many of the lower sections the measured depth was 20 to 21 feet. 
It is known that 96 persons were drowned, and that the losses totaled about $75,000,000. Before the people 





MON. TOE. WED. 


THU. Fill. 


M48l2 46M48l2 4SM48l2 48M48l2 48M48l2 46f 


FT. 

30 

20 

10 







S3 










rts\ 










atp 










ftp 










ftp 








H 


IGh 


W 


AT 


:r 


29 


"T. 










































































































L 


:ve 


e ^ 


HE l( 


»H1 


z 


3 FT. 




























-- 

























































































































































































Diagram III. — Hydrosraph, Great Miami at Dayton, Ohio, March 23-29, 1913. 

could realize the city w r as going to be overwhelmed, the flood wave had swept over the lower portions and 
was rapidly approaching the business center. An hour after the water started over the levees it was impossible 
to cross from the residence sections of the uorth and west iuto the business district, and in less than five 
hours after the water overtopped the lowest points in the levees the entire business section was covered to a 
depth of several feet. The business district and large portions of the residence sections are located in the 
lowlands. This low, flat area, covering about 7.5 square miles, was all submerged during the flood. It is 
approximately 4 miles in length and 2 miles in width. Adjacent to this level flat are hills with varying slopes 
having an area of about 9 square miles and used chiefly for residence purposes. The Mad River, Stillwater 
River, and Wolf Creek discharge into the Miami River within the city limits. The levees were constructed 
to safeguard against a stage of 23 feet, which is 6 feet below the crest of the flood of March 25. 

On Sunday morning, March 23, the reading of the river gage was only 3 feet. It was 7 feet on the 
morning of the 24th. and the rise during the day was moderate and fairly steady. At 12 o'clock noon the gage 
reading was 11.6 feet; at 4 p. m. it was 12.2 feet; at 8 p. m. it was 14.2 feet: at 10 p. m. it was 15.3 feet: and 
at 12 midnight it was 16.4 feet. About 2 a. m. of tbe 25th the water had reached the flood stage, or IS feet, 
and the rise during the following eight or nine hours was exceedingly rapid. The water started over the levee 
in the north portion of the city about 5 a. m., and by 9.30 a. m. water several feet in depth with a dangerously 
swift current was flowing through the heart of the city. The flood crest was nearly reached by 11 a. m. 
The rise then became very slow, the stage being 28.3 feet about 3 p. m. and reaching a maximum of 29 feet 
about midnight. The water receded slowly, the total fall to 4.30 a. in. of the 26th being 0.7 foot. At 3 p. m. 
the water was 4 feet below the crest, and the fall was fairly steady afterwards, as is shown graphically by 
the curve which is plotted from the several special measurements made during the flood. (See Diagram 
III.) The water lowered to the flood stage by about 10 p. m. of the 27th. and it had fallen below the streets 
in the business section on the morning of the 28th, so 1hat many people were permitted by the military 
authorities to enter the city and start the work of rehabilitation. 



RAINFALL. 



Rainfall contributory to the flood started on Sunday, March 23. On that date there was 
0.48 inch at Dayton, which fell from 7.42 a. m. to 10.56 a. m., a brisk shower occurring between 



52 THE FLOODS OF 1913. 

8 and 9 a. m. Then there was no rain of any consequence until 7 a. m. of the 24th, when rain 
began and continued until after midnight of the -25th. During this period of rain there were four 
heavy showers at Dayton. From 7.10 a. m. to 7.38 a. m. of the 24th, 0.52 inch fell ; from 3.21 p. m. 
to 3.47 p. m. of the 24th, 0.54 inch fell: nearly a half inch fell between 6 p. m. and 7 p. m. of the 
same date, and 0.58 inch fell between 2 a. m. and 3 a. m. of the 25th. From the beginning of the 
rain, at 7 a. m. of the 24th, to about 4.30 p. m. of the 25th, when the river had practically reached 
its crest, the total amount of the rainfall, as shown by the automatic record at the local office, was 
5.17 inches. The amount falling during about the same period at the cooperative station in 
the edge of the city, where the rain gage has a more suitable exposure, was 6.19 inches; 4.40 
inches of rain fell during the 24 hours ending about 7 a. m. of the 25th, and the rainfall during 
the remainder of the day was comparatively light. The rainfall at the cooperative station, 
measured by Mrs. Edith E. L. Boyer, an excellent observer, was about 15 per cent greater than 
the amounts measured at the local office. If 15 per cent is added to the 24-hour fall of 4.40 
inches, the amount is raised to about 5 inches. It is interesting to note that about one-half 
of the 24-hour fall of 4.40 inches was recorded in the four showers, which altogether covered 
about 2 hours. The rainfall at Piqua. about 33 miles north of Dayton, on the Miami River, was 
very heavy. The amount measured there for the 17 hours beginning at 7 a. m. of the 24th and 
ending at midnight was 5.43 inches. 

An examination of the rainfall table (see page 54) will show that heavy amounts were 
recorded on the 24th and 25th at all stations above Dayton, in the drainage area of the Miami 
River and its tributaries. Judging from the available records of rainfall and the time the high 
water reached Dayton, it is believed the average 24-hour rainfall for the entire basin above 
Dayton would be about 5 inches had measurements been made at the several stations for the 
24 hours ending about 7 a. m. of the 25th. The average in the basin for the entire period of 
rainfall, beginning on the 21st and ending on the 27th, was about 8.80 inches. The rainfall was 
not unusually heavy for short periods of time. On May 12, 1886, wlien some damage Avas 
done by floods at Dayton, 4.50 inches of rain fell in about two hours. However, the rainfall 
during the recent flood was the greatest on record falling over the entire basin in that period 
of time. It was also the greatest amount that has fallen in the basin during a period of 24 
hours. The rainfall approaching more nearly the amounts of March, 1913, than any other on 
record was on October 5-6, 1910, when the average for the two days in the basin above Dayton 
was about 5.50 inches. At that time the river only reached 16.1 feet, but the soil was not 
saturated to begin with, as was the case in March. 1913. The most destructive flood known 
before the recent one was on September 17. 1866, but the conditions then were so> different 
from those prevailing now that no comparison can be made. The cultivation of the soil and the 
drainage systems which are now in use provide for the rapid flow of the water into the channels. 
The investigation in the Miami Basin, which has been conducted by noted engineers employed 
by the Dayton citizens' relief committee, has disclosed the fact that the maximum discharge 
of the river at Dayton on March 25 was approximately 250.000 cubic feet per second. As the 
drainage basin of the river is about 2,500 square miles, this would give the unusually large 
average maximum discharge in the basin of 100 cubic feet per second per square mile. The 
maximum capacity of the channel at the time of the flood is figured at 100.000 second-feet. This 
would leave 150,000 second-feet flowing through the city outside the river banks. 

The flood converted the city into a river about 2 miles wide, with a current which was 
generally too swift to operate rowboats. (See chart No. 10, overflowed area in Dayton, Ohio, 
p. 51.) People were marooned in the second stories throughout the flooded districts, and in 
many cases it was necessary to find refuge in the attic or upon the roof. There many of them 
remained from two to three days without food and water. The fires which burned business 
blocks and other buildings added to the suffering. Rowboats were in constant use where it was 
possible to reach the people. The local office began sending out flood warnings at 12.30 p. m. 
of the 24th, first into the most exposed places and later to other districts. The warnings were 
continued until about midnight, and it is believed they resulted in the saving of many lives. 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 53 

A number of families moved from low places on the evening of the 24th, and others were 
keeping watch on the river. About 2.30 a. m. of the 25th, when the rise in the river was 
increasing and the rainfall was alarmingly heavy, whistles were blown and bells were rung to 
warn the people of the danger. About 3 a. m. messengers started through the lower residence 
sections asking the people to leave their homes, and this was continued until the water flooded 
the streets. The loss of life was exceedingly small when compared with the severity of the 
flood. 

Portable structures and the lighter frame houses were swept away, and many founda- 
tions of substantial buildings were undermined. Several miles of asphalt pavement were 
ripped from the streets and broken to pieces. It would be difficult to describe with any degree of 
accuracy the enormous financial losses resulting from the flood. The most authentic da.ta as 
to the losses sustained were secured by the Dayton citizens' relief committee after a careful 
investigation of all interests and personal inspection of 2,164 residences in the flooded zone. 
This report is as follows: 

Loss to public property $2,008,100 

Loss to public utilities, steam, street, and interurban, gas and electric lighting companies, tele- 
phone and telegraph companies 5,884,573 

Loss to public utilities, account of loss of business J 838, 631 

Fire loss over insurance 975, 236 

Damage to buildings 15, 200, 000 

Damage to household furniture and furnishings 9, 440, 000 

Loss to merchants on stock and fixtures 18,000,000 

Loss on live stock, automobiles, and vehicles 1,000,000 

Factory losses : 

Wages 4, 045, 000 

Stock and machinery 8. 747, 500 

Business loss . 1, 900, 000 

Loss on contracts, rents, etc 3,450,000 

Pianos in homes 800, 000 

Leaf tobacco in warehouses 900,000 

Total 73, 249, 040 

DAYTON FLOODS IN PAST YEARS. 

The first severe flood that we have record of at Dayton was in March, 1805, which is said to 
have been clue to the thawing of deep snows and the occurrence of heavy rains in the headwaters 
of the Miami and Mad Rivers. During that flood the water covered the whole town, except a 
portion of the business center. According to local history, the people seriousty considered the 
proposition of moving the town from the lower flat to the higher ground in the east portion. 
The next flood of importance was in 1814, when the water was said to have been deep enough at 
the corner of First Street and the canal to swim a horse. The third notable flood occurred on 
January 8, 1828, when a warehouse was washed away from the head of Wilkinson Street and 
the entire southern part of town was submerged. Another flood visited Dayton in February, 
1832. The high-water mark at that time equaled the high-water mark of the flood in 1828. 
A flood occurred on January 2, 1847, which covered everything except an island in the center 
of town. The water came up to Fifth Street, but it was not so high as the flood of 1805. The 
loss was $5,000. The most severe flood of record, with the exception of the flood of March 25, 
1913, occurred on September 17, 1866, when, it is stated, heavy rains for three clays caused a 
rise in the river which overflowed practically the entire town. The water was 1 foot deep on 
the floor of the Beckel House and 4 inches deep on the floor of the Phillips Hotel. The loss 
was $250,000. As the town then had only about 15,000 inhabitants, this loss was comparatively 
great. The next flood was on February 3, 4, and 5, 1883. The danger at that time was 
increased by large ice gorges which formed in the stream. The water came up as high as 
the flood of 1847, but it was about 2 feet below the high-water mark of 1866. Wolf Creek 



54 



THE FLOODS OF 1913. 



rose to an unprecedented height, and the western and southern parts of the city were under 
water. On May 12, 1866, another flood occurred. At that time 4.50 inches of rain fell at 
Dayton in about two hours. Wolf Creek was high, and most of the damage was done by the 
overflow on the west side and the back water in the south portion. In March, 1897, and March, 
1898, severe floods occurred, the latter being the most destructive. Much damage was done 
in North Dayton and Eiverdale by these floods. During the flood of 1898 work Avas done on 
the levee to save the business section, and it is stated that a rise of 6 inches more would have 
flooded that district. 

Rainfall over the Great Miami Hirer watershed above Dayton on Mar. 2J t and 25, 1913. -when the l/caricst 

rainfall occurred. 



Stations. 



Bellefontaine 

Dayton 

Dayton (cooperative) 

Greenville 

New Bremen 

Piqua 

Plattsburg 



Mar. 24. 


Mar. 25. 


Inches. 


Inches. 


1.52 


5.61 


2.95 


2.27 


2.91 


3.28 


1.77 


4.45 


1.80 


3.22 


5.59 


U.71 


1.75 


2.01 



Stations. 



Sidney 

Springfjeld . . . 

Urbana 

Wapakoneta. . 

Average 



Mar. 24. 



Inches. 
1.84 
2.01 
2.13 
1.66 



2.36 



Mar. 25. 



Inches. 
3.96 
3.57 
3.12 
3.29 



3.32 



i To midnight of the 24th. 
Rainfall over the Great Miami watershed above Dayton for the period from Oct. .'/ to 7, 1910. 



Stations. 


Oct. 3. 


Oct. 4. 


Oct. 5. 


Oct. 6. 


Oct. 7. 




Inches. 


Inches. 
0.06 
.07 
.70 
.14 
.89 


Inches. 
3.82 
2.65 
2.50 
2.27 
1.87 
2.80 
3.00 
3.25 
2.80 


Inches. 
2.78 
2.98 
1.94 
2.86 
2.24 
1.84 
3.72 
2.46 
3.50 


Inches. 






10 










0.04 


. 10 










2.20 






T. 

.27 














.10 














.24 


2.77 


2.70 


.28 









Note. — On Oct. 7, 1910, the river reached 16.1 feet. 



Miamisburg is below Da} r ton. Only a small portion of the village was affected by the flood; 
but 1 life was lost and 16 buildings wrecked. 

Franklin is the next town down the river, and the part of the resK 1 *nce section that is on 
the west bank of the river was flooded, and much damage was done, fourteen buildings were 
washed away and 7 people drowned. 

Middletown is in northeastern Butler County and has a population of about 15,000. About 
25 per cent of the city was flooded, including the business section and the better residence 
portion. Six people were drowned and 41 houses destroyed. 

The city of Hamilton is in south-central Butler County and has a population of 39,000. 
It is built on both sides of the Great Miami River, with the main business and residential 
portions on the east bank. The land is very flat, and the main built-up portion of the city lies 
in the level valley of the river. 

The official river gage at Hamilton showed 4.8 feet of water in the river on the morning of 
the 24th. By noon it had risen 4 feet, and at 5 p. m. the reading was 12 feet. At 6 a. m., 
March 25. the reading was 18.8 feet, at 5 p. m. it was 25 feet, and at 3 a. m. of the 26th it had 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 55 

reached its highest point of 34.6 feet. The observer, Mr. Earl W. Stout, reports that by noon 
of the 25th the water was 2i feet over the Main Street Bridge and that the bridge was destroyed 
at 12.25 p. m. Later in the day the railroad bridges went out, thus leaving no connection with 
the two sections of the city. By the night of the 25th the water covered all streets on the east 
side of the river and four streets running north and south on the west side. 

The water was not high enough in most of the business section to damage goods on the 
first floor, but in some of the residence districts it was from 10 to 18 feet in depth. 

Forty-six per cent of the city was flooded. 335 houses were destroyed, and 98 lives were lost. 
The following statement of loss has been furnished by the secretary of the citizens' relief 
committee, dated May 16, 1913 : 

Loss of life, probably 150 

Bodies recovered to date 92 

The city of Hamilton $404,300 

Public schools 2, 500 

County public works 667. 000 

Coke Otto suburb 60,000 

Public communication utilities 105, 500 

Farm lauds 425,000 

Railway, steam and electric 5S5, 000 

Factory losses 2, 799, 218 

Residence lands 134, 955 

Residence buildings 1, 476, 480 

Household goods 1. 565, 770 

Mercantile stock 1, 453, 090 

Live stock 13, 988 

Public library 21, 000 

Total loss . 9, 723, 801 

The above estimate does not take account of depreciation in value of real estate of the city of Hamilton, 
which depreciation will be a net loss unless the city can be proteced from the danger of future inundation. 
The consensus of opinion among all real estate men is that this depreciation amounts to about 33J per cent. 
The total taxable value of the real estate of the city of Hamilton was $31,838,420. The loss through deprecia- 
tion, based on the theory that flood protection is not to be provided, amounts to $10,612,S06. The aggregate 
losses, direct and indirect, will therefore be $20,336,607. It is our opinion that real estate values will quickly 
recover when adequate protective works have been provided. (C. R. Green, secretary.) 

THE LITTLE MIAMI. 

The Little Miami watershed has a drainage area of about 1,709 square miles. The river rises 
in southern Clarke County, flows southwesterly near the western edge of the watershed, and 
empties into the Ohio Biver just above Cincinnati. Its principal tributaries all come from the 
east. The grade is steep in the upper part of the river ; in the lower part the valley is narrow 
and the adjacent bluffs comparatively high. 

At Oregonia, in Warren County, the Weather Bureau river gage showed 7 feet on the 24th 
and 18 feet on the 25th. It reached 21.1 feet at about 11 p. m. of the 25th. Water was in every 
house in the small village, and considerable damage resulted. 

Another river gage is located at Kings Mills, in southern Warren County. The river was 
only 3.3 feet on the morning of the 24th. It had risen to 13.8 feet at 3 p. m., to 17.8 feet at 
7 a. m., March 25, 22 feet at 2 p. m., and 33.7 feet by the morning of the 26th. This was 6.5 
feet higher than during the flood of March, 1897. The observer reports that the river rose 
steadily on the 25th till about 4.30 p. m., when it came to a stand and started falling. Soon 
after 5 p. m. the river began rising with great rapidity, and from 6 to 10 p. m. the rise was 
so rapid that little could be done to save property. The town of Kings Mills is located on the 
high bluff far above the river, but the mills of a large cartridge and powder factory are down 
in the narrow valley and sustained considerable damage. 



56 THE FLOODS OF 1913. 

Morrow is in the narrow valley at the month of a large eastern branch called Toclds Fork. 
The population is less than 1,000 in number, but practically all of the residence and business 
section of the town was flooded. The water was from 7 to 10 feet deep in the stores on Main 
Street and much damage was done to the goods. As the current was at no time very strong the 
buildings were not seriously damaged. 

South Lebanon, with a population of 670, suffered a good deal of damage. The valley here 
is about three-fourths mile wide, with high hills closing in and making the valley much nar- 
rower below the village. Practically the entire settled portion of the town was under water, 
the depth in some of the business streets being fully 25 feet. There is quite a fall through 
the town and a number of houses were wrecked by the rapid current. All stock in stores and 
much furniture was destroyed. 

Loveland. farther down the valley, has a population of 1,476. The entire business district 
was flooded to the depth of 5 to 10 feet, and about 25 per cent of the residential area was affected 
by the water. 

SCIOTO RIVER. 

The watershed of the Scioto River covers 6,301 square miles. It rises in the eastern portion 
of Auglaize County and empties into the Ohio River at Portsmouth. The river is 190 miles in 
length, and has an average fall from source to mouth of 2.6 feet per mile. The principal tribu- 
taries are the Little Scioto, Olentangy, Big Walnut, Big Darby Rivers, and Deer and Paint 
Creeks. 

The rainfall was so heavy that damage was done on all of the small streams Avhich go to 
make up the Scioto River, Kenton, Larue, Marysville, and Marion all suffered much damage 
because of the flooding of the residences. 

The first point on the Scioto River at which a river gage station is located is Prospect, in 
southern Marion County. This place has a population of about 900 and 75 per cent of the town 
was flooded. The water was 5 feet deep in the main street. No lives were lost there and no 
houses were destroyed. The water was 8.5 feet there at 3.30 p. m. March 24, the river having 
risen 3 feet at that time. It rose to 20 feet on the 26th, reaching 4.5 feet higher than during 
the flood in 1904. 

The first station on the Olentangy River is at Delaware, Delaware County. The river there 
was at 5.5 feet on the morning of the 23d and the same on the 24th. It began to rise during 
the forenoon of the 24th, and by 1.30 p. m. was 8.5 feet. It continued to rise with remarkable 
rapidity during the afternoon. At 10 p. m. the gage showed 13.8 feet; 11 p. m., 14.5 feet; 
midnight. 14.9 feet; 1 a. m., 16.8 feet; and at 2 a. m. the 25th it was 18.2 feet. It was not 
possible to reach the gage after 2 a. m., and by morning the bridge on which the gage was set 
was washed away. Later levelings, however, show that the water reached 27.5 feet at noon on 
the 25th, or 1 1.2 feet higher than the highest before recorded, on April 1, 1904. 

The river bed at Delaware has been steadily narrowed by bridges, railroad fills, and fills 
for manufacturing sites, so that it is 180 feet narrower than it was in 1870. In the southern 
part of the city a high railroad embankment crosses the valley, and the special river observer 
reports that the water was from 6 to 10 feet higher above the bridge than below it. 

Only about 12 per cent of the city was flooded, but the water came in the night and rose 
so rapidly that there were 18 deaths from the flood. Twenty-three homes were destroyed, 34 
were partially wrecked, and business firms suffered a severe loss. All of the bridges across the 
river were carried away. 

Four and one-half miles above the confluence of the Scioto and Olentangy Rivers at the 
northern edge of the city of Columbus a large concrete dam has been constructed for water- 
storage purposes. This reservoir has a capacity of 230,000,000 cubic feet, and the watershed 
area above the dam is 1,032 square miles. This so-called "storage dam" is slightly over 1,000 
feet in length and consists of two abutment sections, one at each end, and an overflow section 
between. The overflow section is 500 feet in length, and was designed to discharge a maximum 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 57 

rainfall of 6 inches on the drainage area, flowing off in 24 hours, or about L66,500 cubic feet per 
second. 

The gage at the storage dam showed 1.8 feet of water running over the overflow section 
at 7 a. m. March 24. It rose 1 foot during the next 4 hours. The water rose 1 foot from 2 to 
4 p. m. on the 24th and 1 foot from 9 to 10 a. m. on the 25th. It reached its highest point of 
12.8 feet at 2.30 p. m. on the 25th. 

Columbus is in Franklin County, at the confluence of the Scioto and Olentangy Kivers. 
The main portion of the city is on the east bank, from 50 to 100 feet above the river. The 
Scioto River flows in from the north and strikes the city limits at the northwest corner of 
the western extension of the city. It flows easterly for about 2 miles, then southerly in a 
winding course through the city. The Olentangy River comes from the north along 
the western border of the northern extension of the city and unites with the Scioto just before 
it makes the southerly bend. The total area of that part of the city west of the Scioto River is 
4.37 square miles, and 2.7 square miles of this is in the low level flood plan within the bend. 
The elevation is from 2 to 18 feet above the low-water stage. 

This plain is protected by levees on the southern and western bank of the Scioto River, 
extending a distance of about 7 miles. It is about 2.5 miles from east to west and 1 mile from 
north to south. The northern portion of this district is given over to railroad yards and there 
are some factories and business houses, but most of the area was covered with residences. At 
the beginning of the flood week the population in this section was about 22,500. The river was 
crossed by 15 bridges within the city limits. 

The Weather Bureau river gage is located on the abutment of the most southern bridge 
after the river has made its long curve through the city. On March 23, 1898, the river was 
up to 21.3 feet on this gage; the west side levees were overtopped in places and considerable 
damage resulted. After this flood the leA T ees were strengthened and raised 6 feet through- 
out the whole distance of 7 miles. The estimated discharge of the two rivers during this 
flood was 75,000 second-feet, or 47.9 second-feet per square mile of watershed area. It was 
assumed that the addition of 6 feet to the levees would furnish ample protection against 
future floods. About one-half of the levee, or that part along the southern bank of the Scioto 
where it makes the easterly curve, has been occupied by railroad tracks in daily use. Within 
the last few years the railroad tracks crossing this low plain have been raised to eliminate 
the grade crossings. The Pennsylvania line parallels the river running from west to east and is 
a short distance south of the levees, the Baltimore & Ohio track runs across the western portion 
of the district from southwest to northeast, and two other roads cross it from north to south 
near the eastern limit. These road embankments are pierced by numerous subways at the street 
intersections, and it was the rush of water through these subways that caused the greatest 
destruction of buildings and the greatest loss of life. 

The flood stage in Columbus is 17 feet on the gage, when the water begins to flood buildings 
and low streets on the east side of the river Avhere not protected by levees. The river was 6.2 
feet at Columbus on the morning of March 24. It rose steadily through the day and was 13.8 
feet by 4 p. m. It passed the flood stage soon after 9 p. m., and by daylight of the 25th 
had passed the 1898 record. It continued to rise slowly at the gage until it reached 22.9 feet 
at 11 a. m., stood nearly stationary for an hour, and then fell slowly. It stayed above the flood 
stage until about noon of the 28th. 

These figures show that the water at the Weather Bureau gage w T as only 1.6 feet higher 
than in 1898, although a 6-foot higher levee had been overtopped and washed out. The fact 
is that the water was held back by the bridges that cross it within the city limits and the river 
was from 6 to 7 feet higher above the confluence of the two streams than in 1898. There were 
eight bridges between the mouth of the Olentangy River and the Mound Street Bridge, where 
the gage is located. The narrowest was the Broad Street Bridge, just after the turn is made 
southward, which had an opening only 290 feet wide at right angles to the thread of the 
current. 



58 . THE FLOODS OF 1913. 

Mi*. Julian Griggs, former city engineer of Columbus, states that when the storage dam 
showed a head of 10 feet passing over it at 8 a. m. on the 25th it was carrying the river channel 
capacity of 63,246 second-feet, computing by the Francis formula with 4 as a constant for a 
round-topped weir. At its crest the storage dam was passing 46 per cent more than the channel 
capacity and there was more than the channel capacity going over the dam until 10 a. m., 
March 26. 

The effect of the flood was felt first along the east side of the Scioto and along the Olentangy 
from the northern limit of the city southward to the Scioto. 

The water reached the top of the levee and began to flow over into the west side district 
at 9.25 a. m., March 25. In a short time the embankment had been overtopped at a number 
of places and the water poured over the southern bank of the Scioto above the Olentangy into 
a basin about 1 mile long and one-fourth mile wide between the river levees and the railroad 
embankment. The water then passed through the three main subways under this elevated 
track with a tremenduous scouring effect. There were six gaps in the levees aggregating in 
length 3,300 feet. 

At the crossing of the Little Miami Eailroad which parallels the river and the Baltimore 
& Ohio Eailroad which crosses the western part of the district, 100 feet of the embankment 
was washed out, and at one of the subways one of the abutments was washed out and the 
embankment behind it for a distance of 150 feet. Buildings that were in line with the currents 
through these breaks were completely destroyed. 

After passing the western part of the Little Miami Bailroad the water piled up west of 
the Baltimore & Ohio tracks, rising in a few minutes to a depth of 17 feet. This carried the 
water over the second-story floors in a populous and desirable residence district, either driving 
the inhabitants to the attics or to the roofs. 

The water rushed through subways of the Baltimore & Ohio and meeting other currents 
flowing southward on the east side of the track they formed whirlpools and gave a scouring 
effect that no buildings could withstand. It was in this vicinity that the greatest damage was 
done to residences and the greatest loss of life occurred. Two streets, Glenwood and Central 
Avenues, were closely built up with substantial frame houses, but after the flood practically 
every house for a distance of four squares was gone. Only fragments of foundations were 
left where before there was a happy and prosperous neighborhood. One house from Central 
Avenue was found 4 miles below the city. This damage was done during the evening and 
night of the 25th, fully 12 hours after the levees were overtopped. These people had been 
warned in ample season to have escaped to higher ground, but no one realized that this district, 
protected as it was by a high railroad embankment, would get more than a moderate amount of 
back water. 

Measurements by the city engineer show that the water was 3.04 feet above the curb at 
Broad and Sandusky Streets ; 9.96 feet at Broad and Glenwood ; 8.73 feet at Rich and Cypress ; 
9.66 at Mound and Glenwood; and about 17 feet at West Park and Sullivant Avenue. 

The work of rescue went on as rapidly as the depth of water and swiftness of the currents 
would allow, but it was not until the 28th that the water had gone down sufficiently to allow 
teams to go through the streets. There was much suffering as a result, as many people were 
shut in without relief from Tuesday until Friday. There were many thrilling acts of rescue 
and many heartbreaking stories of inability to save loved ones perhaps even after escape from 
wrecked houses. Thirteen people Avere rescued from one tree alone. 

There were 93 lives lost in the flood in Columbus and the deaths indirectly clue to the flood 
bring the number up to at least 100. According to the record of the city building inspector 
293 dwellings and business houses were totally destroyed, their valuation being $433,700. There 
were 4,071 damaged; the loss on these being $814,200. In 3,100 houses visited after the flood 
it has been estimated that the loss on pianos alone in Columbus amounts to $630,000; loss to 
furniture including musical instruments. $1,320,000; and loss to buildings and other real estate 
$2,250,000. It has not been possible to ascertain the mercantile loss, but a conservative estimate 
makes the total damage to West Columbus by the flood between $12,000,000 and $15,000,000. 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 59 

The city street cleaning department removed 70,000 loads of debris at a cosl of $100,000. 
The repairs to sewers cost $25,000, and the repairs to streets $'.'1,000. The loss of city bridges 
amounts to $350,000. The loss to the county in bridges and highways was $1,200,000. 

Below Columbus no very great damage was clone by the high water until Chillicothe was 
reached. At Circleville, 27 miles below Columbus, the water reached 21.2 feet at 3 a. m.. 
March 26, which was 4.9 feet above any previous record. The city itself is at a good elevation 
above the river and little damage was done. 

Chillicothe. Ross County, is located at the confluence of Paint Creek with the Scioto 
River. The main portion of the city is built on moderately low and level land and slope- away 
from the Scioto River toward Paint. Creek. Chillicothe is 25 miles below Circleville. The 
water was at 11.9 feet at 7 a. m. of March 25. At •_' p. m. it had risen only to 13.4 feet and 
was rising so slowly that the observer did not look at the gage when he left his work in the 
evening. At 2 a. m. of the 26th it had risen to 18 feet and by 6 a. m. had reached 35 feet. Its 
highest point was 37.8 feet at 11 a. m. 

The water broke through the railroad levee at about 5.15 a. m. and about the same time 
was backing up through the sewers and doing 1 some flooding. The railroad levee gave way 
when the water was at its highest point and fully 75 per cent of the city was flooded. The 
water in the east of the city was only about 1 foot higher than in 1881 but it was fully 7 feet 
higher in the west end, caused by the railroad line backing it up. At the river gage the water 
was 9.5 feet higher than the 1898 flood. 

The greatest damage at Chillicothe was clone in Hickory Street south of Main Street, 
where the water plowed out a channel from 6 to 10 feet deep the full width of the street, and 
in some places the houses and property abutting thereon. There were 18 deaths in Chillicothe 
due to the flood. 

Small villages along the Scioto below Chillicothe were flooded and many houses washed 
away. The Norfolk & Western triple tracks at Waverly, Ohio, were twisted and wound into a 
corkscrew by the flood. 

The city of Portsmouth, at the mouth of the Scioto River, had 75 per cent of its area, 
under water. It is stated that when the Scioto flood was at its height it carried a current of 
Avater clear across the Ohio River that was 5 or 6 feet higher than the level of the Ohio above 
the current. The water was 8 to 10 feet in depth in some of the business streets in Portsmouth, 
and 4,500 residences were flooded. 

MUSKINGUM RTVER. 

The drainage area of the Muskingum River is approximately 7,693 square miles, or nearly 
one-fifth of the area of the State. The distance from the mouth at Marietta to the headwaters 
of the Tuscarawas River is 211 miles. 

The northern part of the watershed is in the glaciated portion of the State and is composed 
of smooth, rolling plains cut into by broad stream valleys. From Muskingum County south, 
however, the land was not subjected to glacial action and the streams have cut deep, narrow 
channels. The river is navigable from Zanesville to the mouth, a distance of 70 miles. The 
drainage area above Zanesville is close to 6,500 square miles. The river at Zanesville is 450 feet 
wide at average low water. 

The main branches of the Muskingum River are the Licking River, which unites with it at 
Zanesville from the west : the Wills Creek, which flows in from the east between Zanesville 
and Coshocton, and the Walhoncling and Tuscarawas, which unite to form the Muskingum at 
Coshocton. The Licking River rises in the Licking Reservoir, which covers 3,500 acres and 
drains 90 square miles. The Tuscarawas River at Coshocton is 300 feet wide and the Wal- 
honding 100 feet. The two rivers drain 4.600 square miles. 

The rainfall was very heavy in Richland, Ashland, and Wayne Counties, all of the small 
streams at the headwaters of the larger rivers were higher than ever before known, and a 
great deal of damage was done. At Clinton, in Summit County, for example, near the head- 
waters of the Tuscarawas River, the business section was inundated by about 6 feet of water, 
and great damage was done. 



60 THE FLOODS OF 1913. 

At Massillon, in Stark County, a city of over 14,000. fully 30 per cent of the city was flooded. 
Thirty per cent of Mount Vernon. Knox County, was flooded; 90 per cent of Warsaw and of 
Xellie, both small places in Coshocton County; 50 per cent of Dresden, in Muskingum County; 
and 33 per cent of Belleville, in Richland County. 

At New Comerstown, in Tuscarawas County, which has about r>00 houses, 200 were in the 
flooded district, the water being from -1 to 5 feet in depth. 

The headwaters of the Walhonding River began to rise on the 24th and reached their 
highest point on the 25th, while the Tuscarawas River did not begin to show much change 
until the 25th. Thus at Mount Vernon, Knox County, the highest stage Avas reached at noon 
of the 25th, and at Warsaw, just above Coshocton, the highest was at 5.30 p. m. of the 25th. 

In the Tuscarawas the flood crest did not reach Massillon, Stark County, until 8 a, m. of 
the 26th and New Philadelphia, Tuscarawas County, not till 11 a. m. of the 27th. At Canal 
Dover, the highest water, 16.1 feet, was at noon of the 28th, although it stood at 15 feet from 
4 p. m. of the 26th to 9 p. m. of the 27th. 

Coshocton, Coshocton County, is situated on the east bank of the Tuscarawas River, at its 
junction with the Walhonding River. The city is built on the flood plain, which is compara- 
tively low and level. It has a population of about 10,000. The river gage was on one of the 
piers of the railroad bridge that crosses both rivers about 1,500 feet above their confluence. 
The pier stood in the Tuscarawas River. The two streams at that point are about 100 feet apart 
and the intervening space is overflowed with an 8-foot stage of water. 

By the morning of the 25th the Walhonding River had overflowed the lowlands between 
the two rivers and caused a stage of 11 feet on the gage. At 2 p. m. the gage showed 14.5 feet 
and by 5 p. m. of the 25th the water had reached 20 feet, most of the water coming from the 
Walhonding River. 

No more readings were possible, and before the flood subsided the bridge and river gage 
were Avashed away, 250 residences were flooded, 16 houses were Avashed aAvay, and 4 people 
Avere drowned. The water rose to the second floor in 40 or 50 of the houses. Subsequent 
measurements determined that the highest Avater was 27.5 feet at 3 or 4 a. m. of the 26th. This 
was 5.5 feet higher than any previous record. 

Approximately one-half of the village of Dresden, situated about 15 miles below Coshocton 
and containing some 1,600 inhabitants, was under water. The river here was 11.6 feet higher 
than in 1898. Several houses were Avashed away. 

Zanesville is situated on the Muskingum RiA^er at the confluence of the Licking River. 
It has a population of about 29,000. Part of the city, including most of the business district, 
lies on the east bank of the Muskingum River, and another part west of the Muskingum, 
between that river and the Licking RiA^er. 

The Licking River began rising on the 24th, and caused a rise in the lower pool of the 
Muskingum River from 9.9 feet in the morning to 11 feet by 6 p. m. At 7 a.m. of the 25th the 
gage in the lower pool was at 21.2 feet, and at 6 p. m. it Avas 28 feet, most of the water up to 
this time coming from the Licking River. The Avater began to pour clown thje Muskingum 
River by this time, and) at 2 a. m. of the 26th the Avater at the lower gage was 35 feet, and 
rising at the rate of 10 inches an hour. 

The water reached 39 feet at 6 a. m., 42 feet at 10 a. m., 45 feet at 2 p. m., and 48 feet at 
6 p. m. At 6 a. m. of the 27th it Avas at 49.3 feet, and from 10 p. m, of the 27th until 3 a. m. 
of the 28th it stood at 51.8 feet, 15 feet higher than was registered in 1898 and 17.7 feet higher 
than was reached in 1884. By 7 a. m. of the 28th the Avater had fallen to 50.3 feet and then 
lowered steadily. 

The river gage in the upper pool shoAved 39.1 feet at 6 p. m. of the 27th. Unofficial marks 
show that the water in the upper pool Avas 17 feet higher than in 1898. 

At least 40 per cent of the built-up portion of Zanesville Avas flooded. The Avater was 20 
feet deep at the corner of North and Third Streets. 17 feet at North and Fourth Streets, 15 
feet at Market and Second Streets, and 17 feet at West Main and Luck Streets. 



to 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 61 

The business portion of the city was entirely in the flooded district. Five of the seven 
bridges in the city were carried out. The famous " Y" bridge, a concrete structure, was over- 
topped, but withstood the terrific strain, although part of the superstructure was carried away. 

As the Muskingum Hirer makes a bend to the westward around the business portion of the 
city there was a tendency of the flood waters to cut a new channel across the land through the 
heart of the business portion. The current there was exceedingly swift and destructive. 

An interesting phase of the flood was the damming of the Licking River and the actual 
flooding upstream of this river from the Muskingum when the flood in the latter was at its 
height. The Licking River, which at its own flood crest was 18 inches higher than in 1898, 
Avas buried beneath the water of the Muskingum on the 27th, and the water from the Musk- 
ingum backed up the valley of the Licking for a distance of 9 miles and spread out at two points 
to a width of 21 miles. This spreading out of the Muskingum in this valley, as well as in several 
other smaller ones, was, indeed, fortunate for Zanesville. 

There were 3,111 buildings in Zanesville under water. 157 of which were entirely swept 
away, moved from their foundations or badly wrecked. The losses on buildings and contents 
at Zanesville has been estimated at $2,795,792. The loss to bridges, railroads, and telegraph and 
telephone companies was nearly as much more. 

Zanesville was much more fortunate than many other cities in Ohio, however, because 
there were only 2 lives lost in the flood in this city out of nearly 15,000 which were in the 
flooded district. In one case a woman refused to leave her house when boatmen called to 
rescue her, and the other, a man, had ample time in which to save himself but delav'ed too long. 
There were many daring rescues, however. 

A careful record was kept on Market Street by an interested resident, and he states that 
the river rose there at the rate of 6 inches an hour between 1 and 7 p. m. on the 26th, 1 inches 
an hour between 7 and 9 p. m., 2-| inches an hour from 9 p. m. till midnight, 1 inch an hour 
from midnight until 1.30 a. m. the 27th, and less than three-fourths inch an hour from that 
time until 9 a. m., when the crest was reached. 

It was noticed in a number of cases that small frame houses that had a slate roof were 
overturned in the water and floated bottom up. 

The relief work of the Boy Scouts in Zanesville deserves special mention. Mr. Thomas 
TV. Lewis in his report of the flood states that every boy of the troop was on duty 21 days 
during the flood. 

Immense damage was done in the narrow Muskingum River Valley below Zanesville. 
Farm buildings were destroyed and many acres of rich farming land actually washed away 
or covered with gravel. Out of a total of 31 houses between Zanesville and McConnelsville 
only 13 were left standing. 

At Philo the river stage was 41.9 feet in the upper pool and 50.6 feet in the lower pool, 
16 feet higher than ever before recorded. Bridges and four dwellings were swept away. 

About 25 miles below Zanesville the towns of McConnelsville and Malta are located. 
McConnelsville has a population of nearly 2,000 and is situated on the east bank, and Malta, 
with about 1,000 population, is directly opposite. The river here reached a stage of 10.8 feet, 
or 14.1 feet higher than any previous record. The water reached a depth of 30 feet in the 
parts of McConnelsville nearest the river, and 28 buildings were washed away, but the greater 
part of this town is well up on the hillside out of flood danger. 

Malta, however, is located down near the river, and 60 per cent of its was flooded and the 
property damage was very great. The following notes are furnished by Mr. C. H. Morris, the 
cooperative observer at McConnelsville : 

" No ; there isn't anybody worrying down the river about high water so far as I know." Thus spoke the 
agent on the Baltimore & Ohio northbound train from Marietta on Tuesday night, March 25, which train 
came in one and a half hours late, and which, incidentally, was the last train we saw for nearly three weeks. 
It began raining on Sunday and the rainfall was pretty continuous until Thursday morning, when it 
totaled 4.43 inches. On Tuesday afternoon at 5 o'clock a message came from Zanesville stating that they 
expected there 10 feet more water than had ever before been recorded. We, figuring from our past experience 



62 THE FLOODS OF 1913. 

that the broad bottoms of the Muskingum Valley which had taken care of so much of the flood water hereto- 
fore and would do it again, counted on probably 5 feet more water here than in 1898. Well, we got it, and to 
spare. All day Tuesday the river rose steadily, but not alarmingly, until at dark it began rising in earnest. 
reaching the gage mark of 19 feet at 11 p. m. Then the people along the river front who had decided to await 
that morning for developments got busy, and drays, wagons, and all available persons were requisitioned 
for aid. 

On Wednesday morning by 7 o'clock the gage showed 25 feet, being within a foot and a half of the 1898 
rise, and the angry waters were piling up as never before, for at 10 a. m. they passed the former record of 
2G.4 feet and continued to swell at a greater rate than ever. From now until the middle of the afternoon 
the rale of rise was 13 inches to the hour, and at 9 p. m. the stage of water showed 36.4 feet, which was 
exactly 10 feet in 11 hours. At half past 12 noon the big middle span of the bridge, struck by a heavily 
timbered building, turned gracefully over and sank quietly in the yellow torrent. An hour later the Malta 
span similarly met its fate. 

During the afternoon an unending procession of mills, bridges, railroad cars, barns, dwellings, everything 
that is indispensable to civilization, went pell-mell down the still rising waters, and darkness came again — 
but no sleep. Hundreds of people began making their second move. As the flood swiftly encroached upon 
streets heretofore far removed from the hungry river, and with all gage marks gone, Thursday morning, the 
27th, dawned cold and cheerless with snow in the air. 

The river now had reached a height of 39 feet, and was rising a bare 2 inches per hour, and at 9 p. m. 
touched its highest point, 40.8 feet, where it stood until early Friday morning. All day on Thursday the 
turgid Muskingum was the mecca of the inhabitants of the Twin Cities, now as far removed from each other 
as though in separate States, seeing each other from the hillsides, yet having no communication, and hearing 
nothing from the outside world either. 

Malta was much more sorely hurt than McConnelsville. Not a single business there but was put out of 
commission. The mantel and canning factories each were carried aw r ay, the latter dropping a line of canned 
goods for a good mile. Parts of the main buildings of both the Hoffman tannery and the B. M. plow company 
drifted off. The Eogers Building — one of the landmarks of the valley — having borne the onslaughts of the 
ice and water for almost three-quarters of a century, and having four prosperous places of business within 
its walls, floated off with all its contents. The Elk Eye flouring mill — the largest building in the valley and 
one of oldest — lost its north wing on Thursday forenoon and went sailing away, crushing giant trees and 
leveling a brick house in its wake ere it got out into the river: $10,000 would not cover this one loss. 

Altogether the loss of the two towns and vicinity will reach almost $500,000 in denuded farms, buildings 
destroyed and wrecked, and the unparalleled loss in personal effects. 

The river hills at the upper end of McConnelsville closely approach the river at a point almost midway of 
Malta opposite, where the hill likewise juts well over the bottom land, so that when 30 feet of water is 
reached in the valley the current between the two towns is of tremendous swiftness. It is a curious fact 
that the Muskingum River floods in the past have averaged, with one exception. 14-year intervals, each 
exceeding the other by about 2 feet. 

Beverly and Lowell, both small villages in northern Washington County, were badly 
flooded and much damage was done by the high water. No lives were lost. The water reached 
46.5 feet at Beverly at 2 a. m., March 28, being 15.5 feet higher than ever before recorded. 

Marietta is on the Ohio Biver at the mouth of the Muskingum Biver. The latter stream 
divides the city into two sections and enters the Ohio Biver at right angles. During this flood 
the Muskingum Biver rose rapidly for almost 48 hours before the Ohio Biver reached flood 
stage. The water in the Muskingum Biver was 6 feet higher than during the previous high- 
water mark of 1884, and most of the damage in Marietta was from the Muskingum. 

The territory flooded in Marietta includes an area of 3.5 square miles, which was 66 per 
cent of the total area of the city. The flood covered the entire business portion of the city 
and extended to the Marietta College buildings. Sixty-seven per cent of the total number of 
houses in the city were flooded, with 33 per cent flooded to or above the second floor. Of this 
number 120 were destroyed and about 200 others were so badly damaged that thej^ were unin- 
habitable. Two bridges were carried out near the mouth of the Muskingum Biver in Marietta 
at a loss of $150,000. 

• Some of the small towns on the Ohio Biver were completely submerged by the water, not 
a house showing above the water line. 

Tables of hourly precipitation, daily and special river-gage readings, flood loss by counties, 
etc.. follow : 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 

Table 1. — Hourly rainfall by au tomatic (/ages at the ret/alar Weather Bureau stations in the affected district during the 

Mar. 2S-S6, 1918. 



63 

•period 



Date. 


Rainfall by automatic gages for the hour ending 


l a. in. 


2 a. m. 


3 a. m. 


4 a. in. 


5 a. ni. 


6 a. m. 


7 a. m. 


s a. in. 


9 a. in. 


10a. in. 


ii a. in. 


Noon. 


Mar. 23, 1913. 
Fort Wayne. Ind 
















0.05 
.32 


0.26 
.05 


0. 30 

T. 
.08 
.09 

T. 
.01 
.28 


0. 33 
.06 
.26 
.16 


0. 19 


Indianapolis. Ind 














0.33 


.02 


Toledo, Ohio 














.25 


Davton, Ohio 
















T. 
T. 


.23 




Cincinnati, Ohio 


















Sandusky, Ohio 
















.23 
.16 

T. 


.38 


Columbus, Ohio 


















T. 


.09 


Cleveland, Ohio 


















. 14 


Mar. 24, 1913. 
Fort Wayne, Ind 






0.10 

T. 
.10 
.02 
.01 
.17 


0.30 
.02 
.10 
.01 
.09 

T. 


0.20 
.06 
.03 
.01 
.15 
.04 
.02 


O.OS 
.25 
.09 

T. 
.16 
.08 
.04 












Tndianfipnlis, Tnd 


T. 

0.01 


T. 

0.12 

T. 


.12 
.10 

T. 

T. 
.08 
.03 
.03 


.11 
.12 
.63 

T. 
.02 

T. 
.08 

T. 

T. 


.08 


.02 


.04 


.01 


Toledo, Ohio 




Dayton, Ohio 


.18 
.35 
.09 
.49 
.05 


.IS 
.30 
.07 
. 12 
.14 


.09 
.16 
T. 
.15 
.06 


.08 


Cincinnati, Ohio 




.07 


Sandusky, Ohio 






T. 


Columbus, Ohio 






.26 


Cleveland, Ohio 




T. 


T. 




T. 


Parkersburg, W. Va 












Erie, Pa 










.02 


.02 


T. 


.01 
T. 


.10 
.02 


.07 


.09 


Pittsburgh , Pa 










T. 


Mar. 25, 1913. 
Fort Wayne, Ind 


.10 

.22 

.33 

' .02 


.18 
.18 
.23 
.04 


.02 
.05 
.42 
.58 
.10 
.32 


.02 
.19 
.02 
.21 
.01 
.25 
.06 
.27 


.14 
.36 
.01 
.15 
.09 
.12 
.40 
.24 


.14 
.06 
.18 
.16 
.01 
.12 
.05 
.12 


.02 
.02 
.23 
.30 
.41 
.20 
.07 
.11 

T. 
.17 

T. 


T. 




Indianapolis, Ind 




.04 


T. 


T. 


Toledo, Ohio 


.04 
.24 
.46 
.14 
.18 
.27 
T. 
' .11 


T. 
.06 
.15 
.07 
.13 
.14 




Dayton, Ohio 


.04 
.13 
T. 
.16 
.06 


.20 
.30 

.06 
T. 


. 12 


Cincinnati, Ohio 


.04 


Sandusky, Ohio 


.36 
.01 
.18 


.24 


T. 


Columbus, Ohio 


.15 


Cleveland, Ohio 


.28 


.26 


T. 


Parkersburg, W. Va 


.18 


Erie, Pa 


.13 


.10 


.27 


.25 


.22 


.18 


.10 


.18 


.18 
T. 


.08 


Pittsburgh, Pa 


.OS 


Mar. 26, 1913. 
Toledo, Ohio 


.01 
.11 

T. 
.17 
.05 

T. 
.12 
.25 


.01 
.05 

T. 
.16 

T. 
.04 
.05 
.07 


T. 
.02 
.02 
.04 

T. 
.22 
.08 
.33 
















Cincinnati, Ohio 


.02 

T. 
.03 

T. 
.24 
.02 
.18 












T. 


.03 


.11 


Sandusky, Ohio 














Columbus, Ohio 
















.05 


Cleveland, Ohio 


















Parkersburg, W '. Va 


.18 
T. 
.14 


.09 
.01 
.18 


.02 
.02 
.02 


.01 
.02 


T. 
T. 


T. 


.01 
T. 


.04 


Erie, Pa 


.01 


Pittsburgh, Pa 


















Rainfall by automatic gages for the hour ending — 


Date. 


lp. m. 


2 p. m. 


3 p. m. 


4 p. m. 


5 p. m. 


6 p. m. 


7 p. m. 


8 p. m. 


9 p. m. 


10p.m. 


11p.m. 


Mid- 
night. 


Total. 


Mar. 23, 1913. 
Fort Wayne, Ind. 


0.25 
.01 
.16 


0.28 
.02 
.20 


0.20 
.33 
.28 


0.10 
.11 
.18 


0.11 


0.01 














2.08 












T. 


0.02 


1.27 


Toledo, Ohio 


.20 


.09 


0.20 


T. 






1.90 


Dayton, Ohio 










.48 


Cincinnati, Ohio 


























T. 


Sandusky, Ohio 


.30 
T. 

.28 
T. 
T. 


.30 


.29 


.19 


.23 


.12 


.14 


0.01 










2.20 


Columbus, Ohio 










.53 


Cleveland, Ohio 


.31 
.08 
.09 
T. 


.37 


.19 


.21 


.15 


.09 


.20 


T. 








1.94 


Parkersburg, W. Va 








.08 


Erie, Pa 


.17 

.08 


.25 
.10 


.15 
.02 


.08 


.04 


.17 


0.11 


0.13 


0.09 




1.28 


Pittsburgh, Pa 


.20 



64 



THE FLOODS OF 1913. 



Table 1. — Hourly rainfall by automatic gages at the regular Weather Bureau stations in the affected district during the 

period Mar. 23-26, 1912— Continued. 



Date. 



Mar. , 



1913. 



Fort Wayne, Ind 

Indianapolis, Ind 

Toledo, Ohio 

Dayton, Ohio 

Cincinnati, Ohio 

Sandusky, Ohio 

Columbus, Ohio 

Cleveland, Ohio 

Parkersburg, W. Va 

Erie, Pa 

Pittsburgh , Pa 

Mar. 25, 1913 



Fort Wayne, Ind 

Indianapolis, Ind 

Toledo, Ohio 

Dayton, Ohio 

Cincinnati, Ohio 

Sandusky, Ohio 

Columbus, Ohio 

Cleveland, Ohio 

Parkersburg, W. Va 

Erie, Pa 

Pittsburgh, Pa 

Mar. 26, 1913. 

Toledo, Ohio 

Cincinnati, Ohio 

Sandusky, Ohio 

Columbus, Ohio 

Cleveland, Ohio 

Parkersburg, W. Va 

Erie, Pa 

Pittsburgh, Pa 



Rainfall by automatic gages for the hour ending- 



] p.m. 



T. 
0.23 

T. 

.03 
T. 



.02 
T. 
T. 
T. 



.12 
T. 

.07 
T. 

.03 



.02 



.05 
.01 
.04 



2 p.m. 



0.17 



.03 



.02 

.07 



.01 

T. 

T. 
.01 
.04 
.05 



T. 



.01 



.03 



.02 
.01 
.04 



3 p.m. 



0.07 
.16 
.06 
.01 



.09 
.03 
.02 
.05 



.01 

T. 
.01 
.01 

T. 

T. 



.02 



.19 
T. 

.02 
.11 
.02 
.03 
.09 
.02 

T. 

T. 



4 p.m. 



5 p.m. 



0.01 

.17 
.09 
.58 



.18 
T. 

.16 
T. 



.16 

T. 
.01 
.01 
.07 
.02 

T. 
.04 



. 00 



.10 
.07 
.05 
.19 
.12 
.02 
T. 
.01 



0.10 

.08 
.02 
.14 
.67 
.15 
.17 
.07 



.01 
.02 
.02 

i .04 
.03 
.01 
.03 

T. 
.06 



T. 



.10 
.10 
.11 
.04 
.07 
.01 
.12 



6 p.m. 



0.07 
.02 
.05 
.10 
.25 
.06 
.37 
.22 



.02 

.17 
.02 



.18 
T. 

.05 
T. 



.05 
.06 
.11 
.01 



.12 
.05 



' p. m. 



0.20 
.33 
.07 
.49 

T. 
.09 
.10 
.04 



.16 
T. 



.02 

.17 
.01 



1.12 
.02 

.28 
.04 



.04 



8 p.m. 



0.22 
.22 



.03 
.05 
.02 



.19 



.03 



9 p.m. 



0.09 
.18 
.14 
.14 



T. 
T. 
.05 



.23 

.05 
.29 
.21 

T. 

T. 
.01 



.03 
.16 
.14 
.08 
.04 
.21 
.06 
.04 



10 p.m. 



0.07 
.16 
.24 



.01 



.10 



11p.m. 



0.22 

.21 
.19 
T. 



.02 



.21 
.01 
.15 
.11 
.07 
.05 
T. 



.03 

T. 
.06 
.05 
.18 
.12 
.05 
.06 



Mid- 
night. 



0.08 
.21 
.20 
.07 



.20 



T. 



.02 



.21 

T. 
.15 
.17 

T. 
.09 
.21 



.03 
.01 
.06 
.11 
.03 
.09 
.06 
.04 



Total. 



1.98 

2.76 
1.82 
2.95 
2.21 
1.58 
2.14 
1.46 

.05 
1.38 

.72 



1.56 

1.74 

2 2. 27 

4.15 
2.05 
2.89 
2.66 

.80 
.212 

.55 



1.11 
.95 

1.40 
.91 

1.84 
.91 

1.66 



1 Clock on triple register stopped at 4.30 p. m. 



2 Total to 4.30 p. m. 



Table 2. — Average daily rainfall and the total rainfall for the period Mar. 23-27, 1913, arranged by watersheds; also the 
area of the watersheds, and the total fall in cubic feet per square mile. 



Watersheds. 



"Rainfall for 24 hours ending 7 p. m. 



23d. 



24th. 



2.5th. 



26th. 



27th. 



Total. 



Areas of 
water- 
sheds in 
square 
miles. 



Total fall in 
cubic feet per 
square mile. 



Sandusky 

Mahoning 

Great Miami, above Dayton.. 

Great Miami, total 

Little Miami 

Scioto, above Columbus 

Scioto, total 

Muskingum, above Zanesville 
Muskingum, total 



2.02 
1.26 
.92 
.66 
.32 
1.16 
.71 
.66 
.55 



1.56 
1.36 
1.94 
2.22 
2.09 
1.90 
1.52 
1.26 
1.17 



2.92 
2.90 
3.82 
3.82 
2.54 
3.25 
2.48 
2.62 
2.22 



0.91 
1.10 
1.41 
1.50 
2.07 
1.76 
2.00 
1.73 
1.87 



0.81 
.71 
.48 
.42 
.51 
.60 
.62 
.65 
.63 



8.22 
7.33 
8.57 
8.62 
7.53 
8.68 
7.33 
6.92 
6.44 



1,554 
1,025 
2,558 
5,400 
1,709 
1,567 
6,301 
6,509 
7,693 



19, 096, 000 
17, 029, 000 
19, 920, 000 
20, 026, 000 
17, 494, 000 
20, 065, 000 
17, 029, 000 
16,076,000 
14,961,000 



PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 



65 



Table 3. — Daily river-gage readings during the period Mar. 24-28, 1913, and the highest stage reached with the hour and 
date of its occurrence, the flood stage, the highest previous stage and the month and year of its occurrence, and the amount 
in feet and tenths of feet that the stage in 1913 exceeded the highest previous stage. 

[All daily readings were made at 7 a. m., except where marked with a star (*).] 



Stations. 



Sandusky River: 

Upper Sandusky 

Tiffin 

Fremont 

Great Miami: 

Piqua 

Dayton 

Hamilton 

Little Miami: 

Oregonia 

Kings Mills 

Scioto: 

Prospect 

Delaware 

Columbus Reservoir. 

Columbus 

Circleville 

Chillieothe 

Muskingum: 

Canal Dover 

Coshocton 

Zanesville 

McConnelsville 

Beverljj 



Mar. 24. 



Mar. 25. 



16.8 

7.0 
9.4 

8.7 
7.0 
4.8 

7.0 
3.3 

*8.5 
5.5 
1.8 
6.2 



1.6 

2.3 
2.5 
9.9 



*19.0 
12.5 
13.5 

23.7 
24.0 
19.6 

18.0 
17.8 



*27.5 
9.3 
21.9 
11.6 
11.9 

7.0 
11.0 
21.2 



16.6 



Mar. 26. 



17.6 

19.4 

*21.5 

15.0 

28.1 

*34.6 

21.1 
33.7 



10.4 

20.9 

24.2 

*37.8 

13.0 
*27.5 
*39.0 

24.3 



Mar. 27. 



Mar. 28. 



16.0 
16.0 
21.5 

12.0 

22.2 

*25.0 



8.9 
19.7 
20.3 



15.0 



*51.8 
*40.8 



13.5 
12.0 
14.3 

10.3 

15.7 

*19. 2 



6.6 
17.4 
16.2 



16.1 
50.3 

*46.5 



High- 
est 

stage. 



19.0 
19.4 
21.5 

24.0 
29.0 
34.6 

21.1 

33.7 

20.0 
27.5 
12.8 
22.9 
24.2 
37.8 

16.1 
27.5 
51.8 
40.8 
46.5 



Date. 



Hour. 



9 a. m. 

6 a. m. 

About noon. 

10 a. m. 
12 midnight. 

3 a. m. 

11 p. m. 
3.50 a. m. 



Noon. 

2.30 p. m. 

11 a. m. 

3 a. m. 
11 a. m. 

Noon. 

4 a. m. 
10 p.m. 

9 p. m. 
2 a. m. 



Flood 

stago. 



13.0 
7.0 
10.0 

12.0 
18.0 
12.0 

10.0 
17.0 

9.0 

9.0 

17.0 
12.0 
14.5 

9.0 
8.0 
25.0 
15.0 
25.0 



High- 
est, pre- 
vious 
read- 
ing. 



16.0 
11.4 

17.8 

16.0 
21.3 
23.3 

20.4 
27.2 

15.5 
16.3 
5.5 
21.3 
19.3 
28.3 

12.0 
22.0 
36.8 
26.4 
35.0 



Year. 


Month. 


1883 


Feb. 


1904 


Apr. 


1884 


Feb. 


1866 


Sept. 


1898 


Mar. 


1897 


Mar. 


1897 


Mar. 


1904 


Apr 


1904 


Apr. 


1909 


Feb. 


1898 


Mar. 


1884 


July 


1898 


Mar. 


1898 


Mar. 


1898 


Mar. 


1898 


Mar. 


1898 


Mar. 



Excess 
in 1913. 



3.0 
8.0 
3.7 

8.0 
7.7 
11.3 

.7 
6.5 

4.5 
11.2 
7.3 
1.6 
4.9 
9.5 

4.1 
5.5 
15.0 
14.4 
15.5 



Table 4. — Special river gage readings of rivers in Ohio, Mar. 24-27, 1913. 



Date. 


1 a. m. 


2 a. <n. 


3 a. m. 


4 a. m. 


5 a. m. 


6 a. m. 


7 a. m. 


8 a. m. 


9 a. m. 


10 a.m. 


11a.m. 


Noon. 


Sandusky River: 
Tiffin- 
Mar. 24 














7.0 
12.5 
19.4 

9.4 
13.5 












Mar. 25 


















14.0 




Mar. 26 






















Fremont — 

Mar. 24 • 


















10.5 






Mar. 25 














14.7 


15.2 






Mar. 26 


















21.5 


Great Miami River: 
Piqua — 

Mar. 23 


























Mar. 24 














8.7 
23.7 

7.0 
24.0 
28.1 

4.8 
19.6 


8.9 










Mar. 25.. 
















24.0 




24.0 


Dayton — 

Mar. 24 


















11.6 


Mar. 25 






















Mar. 26 
























Hamilton — 

Mar. 24 




















8.8 


Mar. 25 












18.8 


20.5 


21.7 


22.2 


24.8 




Mar. 26.. 






34.6 








Little Miami River: 
King Mills — 
Mar. 24. 












3.3 
17.8 
33.7 










6.3 


Mar. 25... 


















19.5 





























14584°— 13- 



66 



THE FLOODS OF 1913. 

Table 4. — Special river gage readings of rivers in Ohio, Mar. 24-27, 1913 — Continued. 



Date. 


1 a. m. 2 a.m. 


3 a. m. 


4 a. m. 


5 a. m. 


6 a.m. 


7 a.m. 


8 a.m. 


9 a. m. 


10a.m. 


11a.m. 


Noon. 


Scioto River: 
Delaware — 

Mar. 24 














5.5 












Mar.25 


16.8 


18.2 


















27.5 


Columbus Reservoir — 

Mar. 24 










1.8 
9.5 
10.4 
8.9 

6.2 
21.9 
20.9 
19.7 
17.4 
14.7 

11.9 








2.8 
12.5 




Mar. 25 












8.9 




10.7 


11.7 




Mar. 26 














Mar. 27 


















8.4 






Columbus— 

Mar.24 






















Mar. 25 












21.5 
21.0 


21.9 
20.8 


22.2 
20.7 
19.6 


22.6 
20.5 


22.9 
20.2 


22.9 


Mar. 26 


21.5 


21.3 


21.0 


21.0 


21.0 


20. 1 


Mar.27 




Mar. 28 
















17.1 


16.9 


Mar. 29 














14.6 


14.5 




14.2 


Chillicothe— 

Mar. 25 




















Mar. 26 




18.0 








35.0 








37.8 




MuskingumRiver: 
Canal Dover — 

Mar.25 










7.0 
13.0 
15.0 
16.1 

2.5 
11.0 










Mar. 26 
























Mar.27 
























Mar. 28 






















16.1 


Coshocton — 

Mar. 24 
























Mar.25 
























Mar. 26 








27.5 
















Zanesville — 

Mar. 25 












21.2 












Mar. 26 




35.0 








39.0 
49.3 






42.0 






Mar.27 




















Mar. 28 




51.8 






50.3 












McConnelsville — 

Mar. 25 










• 












Mar. 26 ' 












25.0 






26.4 






Mar. 27 ' 













































Date. 


1 p. m. 


2 p. m. 


3 p. m. 


4 p. m. 


5 p. m. 


6 p. m. 


7 p. m. 


8 p. m. 


9 p. m. 


10 p. m. 


11p.m. 


Mid- 
night. 


Sandusky River: 
Tiffin- 
Mar. 24 


7.8 
























Mar. 25 
























Mar. 26 


























Fremont — 

Mar. 24 


11.5 




11.8 
17.5 
21.5 








12.4 












Mar.25 


















Mar. 26 










21.5 














Great Miami River: 
Piqua — 

Mar. 23 










5.0 












Mar. 24 . . 










11.0 






14.0 












24.0 
12.0 


















Dayton — 

Mar. 24 






12.2 








14.2 




15.3 






Mar. 25 














29.0 


Mar. 26 
























Hamilton — 

Mar.24 


9.8 


10.6 


11.0 


11.7 


12.0 
25.0 














Mar. 25 








1 






Mar.26 





















PRECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 
Table 4. — Special river gage readings of rivers in Ohio, Mar. 24-27, 1918 — Continued. 



67 



Date. 1p.m. 


2p.m. 


3 p.m. 


4 p.m. 


5 p. m. 


6 p.m. 


7 p.m. 


8 p.m. 


9 p.m. 


10 p.m. 


11 p.m. 


Mid- 
night. 


Little Miami River: 
Kings Mills— 

Mar.24 10.7 


22.0 


13. 8 




















Mar. 25 




















Mar. 26 





















Scioto River: 
Delaware— 

Mar.24 




8.5 














13.8 


14.5 


14.9 


Mar. 25 
















Columbus Reservoir — 

Mar.24 




3.3 
9.4 


12.8 


4.3 


12.7 
9.3 
7.8 

14.2 
22.0 


4.8 
12.4 
9.3 




6.0 
11.7 




6.4 




6.7 


Mar. 25 


12.7 


Mar. 26 










Mar. 27 




















Columbus- 
Mar. 24 


8.8 
22.2 
20.1 






13.8 
22.0 


15.0 
22.0 


15.5 
22.0 


16.1 
22.2 


16.8 
22.2 


17.6 
22.0 


18.4 
21.7 


18.8 
21.5 


Mar. 25 

Mar. 26 


22.1 
20.1 


22.0 

20.0 


Mar. 27 


18.5 
16.5 












j 




Mar. 28 


16.8 
13.9 


16.7 
13.4 


16.6 
13.9 


16.4 
13.7 














Mar.29 
















Chfflicothe— 

Mar. 25 
















Mar. 26 
























Muskingum River: 
Canal Dover — 

Mar. 25 






















10.6 




Mar. 26 ". 








15.0 
















Mar. 27 
















15.0 






























Coshocton — 

Mar. 24 




































20.0 
















Mar. 26 


















Zanesville — 












28.0 
48.0 












Mar. 26 




45.0 


















Mar. 27 
















51.8 






Mar. 28 
























McConnelsville — 

Mar. 25 






















19.0 




Mar. 26 
























Mar. 27 












40.4 






40.8 

































Table 5. — Monetary loss to highways and bridges, buildings and personal property, cost of cleaning and repairing basements 
and machinery, damage to crops and live stock, and number of lives lost. 



Counties. 



Adams 

Allen 

Ashland... 
Ashtabula. 

Athens 

Auglaise. . . 
Belmont... 

Brown 

Butler 

Carroll 

Champaign 



Damage to 
public 

highways 

and 
bridges. 



$8,000 
81,600 

110,000 
80,000 
30,000 
58,000 
5,500 
15,000 

600,000 



70,000 



Damage to 
buildings 

and 
personal 
property. 



Cost of 
cleaning 

basements, 
repairing 

machinery 
and other 

equipment. 



$73, 850 

100. 000 

Small. 

10,000 

10,000 



$100 

Small. 

2,000 

500 



15, 000, 000 




300,000 




Crops 
destroyed 

or 
damaged. 



Loss of live 
stock. 





$500 





.50.000 




$4,200 

3,200 

1,100 

400 



650 



7,000 

100 

3,100 



Number 
lives 
lost. 



98 




68 



THE FLOODS OF 1913. 



Table 5. — Monetary loss to highways and bridges, buildings and personal property, cost of cleaning and repairing base- 
ments and machinery, damage to crops and live stock, etc. — Continued. 



Clark 

Clermont 

Clinton 

Columbiana . 
Coshocton. .. 

Crawford 

Cuyahoga 

Darke 

Defiance 

Delaware 

Erie 

Fairfield 

Fayette 

Franklin 

Fulton 

Gallia 

Geauga 

Green 

Guernsey 

Hamilton 

Hancock 

Hardin 

Harrison 

Henry 

Highland 

Hocking 

Holmes 

Huron 

Jackson 

Jefferson 

Knox 

Lake 

Lawrence 

Licking 

Logan 

Lorain 

Lucas 

Madison 

Mahoning.. . 

Marion 

Medina 

Meigs 

Mercer 

Miami 

Monroe 

Montgomery. 

Morgan 

Morrow 

Muskingum.. 

Noble 

Ottawa 

Paulding 

Perry 

Pickaway. . . 

Pike 

Portage 

Preble 

Putman 

Richland 

Ross 



Counties. 



Damage to 
public 

highways 

and 
bridges. 



$100, 000 
25,000 



2,700 
200,000 
45,000 
200, 000 



90,000 
424,000 

5,500 
55,000 
15,000 
1,550,000 
20,000 
30,000 
21,000 
40, 000 
10,000 
1.000,000 

9,700 
45, 264 

2,500 
20,000 



Damage to 
buildings 

and 
personal 
property. 



875,000 

200,000 

25,000 

300, 000 



350,000 

355,000 



20,000 

10,000 

15, 000. 000 

Small. 



500 



350, 000 

1,000,000 

5,000 

1,500 

5,000 



Cost of 
cleaning 

basements, 
repairing 

machinery 
and other 

equipment. 



$5,000 
6,000 
5,000 



10.000 



1,000 

5,000 







100, 000 



5,000 
Small. 



Crops 
destroyed 

or 
damaged. 



Small. 

?50,000 

3.000 



10.000 



2,000 

40.000 



Small. 



6,000 
10. 000 



25, 000 
Small. 



Loss of live 
stock. 



si. ooo 

500 

1,000 

Small. 

610 

400 



1,200 

1,300 

3,150 



1,300 

500 

3,800 



213 



10, 400 

725 

7,100 

1,500 

1,000 

300 



Number 
lives 
lost. 



22 





93 
2 



15,600 

39, 500 

52,000 



41,000 

150, 000 

2,000 

10,000 

31,800 

38,500 

65,000 

75,000 

85, 000 

46, 000 

65,000 

75,000 

20,000 

7,600 

550, 000 

7,000 

1,350,000 

175,000 

75,000 

450, 000 



2,000 

25,000 

15,000 

400, 000 

97,700 

42,500 

41,000 

1, 150 

150,000 

314,325 



100. 000 



1,000 



1,000 



20,000 
1,000 



15.000 
500 



5.000 
200 



10,000 
Small. 
30,000 
50,000 
1,000 
125, 000 



3,000 
Small. 
Small. 

5,000 



50,000 



5.000 
7.000 
2.000 

Small. 
1,000 

15. 000 



1,000.000 



2, 000, 000 

12,000 

32, 000, 000 

434, 800 

25,000 

2,700,000 



100.000 



60,000 

1,000 

2. 500, 000 

4,000 



100, 000 



60.000 



250, 000 

30, 000 

10,000 



250,000 



75. 000 



2,000 



50, 000 
350, 000 
200.000 



2,000 
2,000 
10,000 



15,000 
40,000 
100.000 



100,000 
1.340.000 



50,000 
30,000 



25,000 
20.000 



100 
150 

750 




500 
300 



1,075 

900 

600 

Small. 



1.000 

525 



10. 000 



2,800 

25 

50,000 

400 

100 

200 

2,000 

200 

750 

750 

1,000 

12, 025 

5,000 

700 

4.100 

300 

19.200 




60 

97 

4 
3 



PBECIPITATION AND FLOODS IN OHIO, MARCH, 1913. 



69 



Table 5. — Monetary loss to highways and bridges, buildings and personal properly, cost of cleaning and repairing base- 
ments and machinery, damage to crops and live stock, etc. — Continued. 



Counties. 



Damage to 
public 

highways 
. and 
bridges. 



Sandusky . . 

Scioto 

Seneca 

Shelby 

Stark 

Summit 

Trumbull... 
Tuscarawas. 

Union 

Van Wert . . 

Vinton 

Warren 

Washington 

Wayne 

Williams... 

Wood 

Wyandotte. 

Total. 



865, 500 
325, 000 
300, 000 
120, 000 
225. 000 
240. 000 
125, 000 
200, 000 
225,000 
7,000 
100 
300,000 
180, 000 
100, 000 
5,000 
70, 000 
60. 000 



12,031.039 



Damage to 
buildings 

and 
personal 
property. 



1300, 000 



702, 737 

35,000 

300, 000 

100, 000 

000,000 

50,000 

Small. 

25,000 



400,000 

2, 000, 000 

25, 000 



20.000 




78, 072, 387 



Cost of 
cleaning 
basemen Is, 
repairing 
machinery 
and other 
equipment. 



$5, 000 



70, 000 
30, 000 
50, 000 
Small. 



10, 000 


15, 000 


50.000 



2. 000 



5, 000 




3,762.100 



Crops 
destroyed Loss of live 

or stock, 

damaged. 



15. 000 



50, 000 
40, 000 
10, 000 
Small. 



40. 000 



20,000 

100 

50,000 

300,000 

10,000 

Small. 

5,000 



1,412,800 



SI, 800 



1,200 
1,200 

350 
1,200 
3, 500 
8,000 

925 

500 



30,000 

200 

1,250 

Small. 

2,500 

1,100 



234, 953 



Number 
lives 
lost. 





467 



FLOOD IN THE WHITE RIVER OF INDIANA, MARCH, 1913. 



By C. E. Norquest, Acting Section Director. 



The flood of March, 1913, is without a parallel in the history of Indiana. Water stages 
reached were from 2 to 8 feet higher than those recorded in any previous flood; the loss of life 
and property was unprecedented ; thousands were driven from their homes, fleeing for their 
lives; transportation lines were helpless through loss of track and bridges;- telephone and tele- 
graph lines were crippled; communities were cut off from communication with the outside 
word for from 24 to 48 hours ; cities were deprived of light and power by the flooding of power 
plants; isolated towns were threatened with famine; and for a period of three days or more 
the great commercial enterprises of the State were at a standstill. To this stupendous toll of 
loss by flood the White River watershed contributed a large share. The East and West Forks 
of the White River flow through a populous and prosperous section of the State, and many 
thriving towns and cities are situated along their banks and on their tributary streams. It 
follows that flood stages must be attended by property loss, and when the streams rise so much 
higher than could be reasonably expected, in the light of past experience such unusually high 
water must be attended by enormous destruction of property. 

THE WHITE RIVER VALLEY. 

The basin of the White River comprises more than one-third the total area of the State. 
The drainage area is approximately 11,300 square miles, about equally divided between the 
East and West Forks. The valleys of the two forks are of nearly equal length, about 275 miles, 
and their topography is somewhat similar, both valleys being flat and broad and subject to 
overflow ; the West Fork has an average fall of 3 feet per mile, the East Fork a little less. Both 
streams flow through an excellent farming region, and the bottom lands comprise some of the 
most valuable farm land of the section. The East Fork, because it has the less fall, has a greater 
tendency to meander in its course, and also to change its course during high water, a tendency 
that proved particularly destructive to farm land during the recent flood. The tributaries of 
these main streams are generally short and rapid streams that rise quickly after heavy rains 
and subside with equal rapidity. 

THE HEAVY RAINFALL OF MARCH 23-2T, 1913. 

The immediate cause of the great flood of March. 1913, was the unprecedentedly heavy 
rainfall of March 23 to 27. Contributory causes, such as the obstruction or narrowing of 
channels, no doubt added to the disastrous results in some communities. 

The meteorological records for this region show isolated cases where at a single station 
a greater amount of rainfall has occurred in an equal period of time, but never before in the 
history of the State, as far back as records extend, has so great a depth of precipitation occurred 
so uniformly over the entire watershed in so short a period of time. The worst previous flood 
of the White River was that of March, 1904. The rain causing that flood began on March 24 
and continued for two days, heavy rainfall extending generally over the entire watershed. The 
average rainfall for seven stations, about equally distributed over the watershed of the West 

71 



72 



THE FLOODS OF 1913. 



Fork, was 4.97 inches, as against 7.81 inches recorded during the recent flood. The average for 
eight stations distributed over the watershed of the Elast Fork was 4.19 inches for the flood of 
1904, as compared with 8.41 inches for the flood of 1913. It is also worthy of note that at most 
of these stations more rain fell within 24 hours during the 1913 flood than was recorded in 48 
hours during the flood of 1904. This tremendous downpour occurred over a drainage basin 
the soil of which was already well saturated by previous rains, and whose streams were already 
carrying their normal spring flow. 

Stations reporting precipitation in excess of 5 inches for 2J/ consecutive hours (see also Taole 2). 



Station. 


Amount. 


Date. 


Rlnnmmgtnn 


6.56 
7.00 
6.10 
5.59 
6.01 
5.43 
6.66 
6. 10 


Mar. 25 


Columbus 


Do 


Elliston 


Do 


Mauzy 


Do. 


Nashville 


Do. 


Seymour 


Do. 


Shoals 




Washington 1 


Mar. 25 







The average precipitation for the White River watershed, as determined from 20 stations, 
on March 24 was 2.42 inches; on the 25th, 4.05 inches. 

RIVER STAGES. 

The Weather Bureau maintains regular gaging stations on the White River at Anderson, 
Indianapolis, Shoals, and Elliston. At each of these stations the stages reached during the 
flood of 1913 are the highest ever recorded, and from other towns and cities along the rivers it 
is generally reported that the stages reached are the highest ever known. 

Crest stages as compared with previous highest water. 





Flood 
stage. 


Previous highest stage. 


1913 


Station. 


Height. 


Date. 


Highest. 


Date. 


Com- 
pared 
with pre- 
vious 
highest. 


Anderson 

Indianapolis 

Elliston 

Shoals 


9 

12 
21 
20 


18.8 
19.5 
29.6 
34.1 


Mar. 23,1904 
Apr. 1,1904 
Mar. 5, 1897 
Mar. 30,1904 


22.1 
25.7 
31.3 
42.2 


Mar. 25 
...do 

Mar. 27 
Mar. 28 


+3.3 

+ 6.2 
+ 1.7 
+8.1 









The crest of the flood passed along, the upper reaches of the streams during the 25th and 
26th, along the middle courses on the 26th and 27th, and along the lower courses on the 27th 
and 28th. 

The following records of daily gage readings taken at 7 a. m., ninetieth meridian time, show 
admirably the rapid rate at which the rivers rose. The sharp rise caused by the excessive pre- 
cipitation of the night of the 24th-25th is noteworthy. 



Station. 


22d. 


23d. 


21th. 


25th. 


26th. 


27th. 


28th. 


29th. 




4.3 

4.7 


3.S 


ll.S 
11.0 
11.8 
8.8 


17.6 
18.0 
23. S 
21.6 


20.6 
(>) 
27.8 
29.5 


14.0 


10.2 


7.8 






Elliston 


31.3 

37.0 


30.4 
42.2 


2S.6 


Shoals 


7.4 


8.0 


41.7 







The Indianapolis gage was washed away during the night of 25th-26th. 





n 








1 






1 
I 


14 

Iff* 


§■ 




< 


■ft 

[ii'M ! 

i'H 1 f 


x'l 




■h 


i sfegli 1 «J ' 










" P Mb 


'■ 




- : 'i "" 




HI' 








>5 

crt- 

o 

iiJll 
CDCC 

I 111 
HO 



FLOOD IN THE WHITE RIVER OF INDIANA, MARCH, 1913. 73 

FLOOD SUMMARY. 

Frequent moderate rains had been general over the State from the beginning of the month, 
and on the 20th and 21st heavy rains occurred. As a result the soil was pretty thoroughly 
saturated and the streams carrying a considerable volume of water when the storms of the 
23d to 27th set in. 

Rain began in the early morning of the 23d and continued almost unceasingly until the 
26th. when the steady downpour gave way to showery conditions. By the evening of the 24th 
the smaller streams tributary to the White River were overflowing into the lowlands. At 
Elwood, in Madison County, Duck Creek was 30 inches higher than in 1904; one-third of the 
town was inundated, and over 100 families had been rescued from flooded homes. At Broad 
Ripple, a suburb of Indianapolis, White River was spreading over the lowlands, and many 
families were abandoning their homes. Fall Creek was out of its banks in northeastern 
Indianapolis, and families in that part of the citj^ were moving out. Eagle Creek on the west 
side was far beyond its banks, and people were moving to higher ground as rapidly as con- 
veyances could be secured. The White River was out over the bottoms along its upper reaches 
in Randolph and Madison Counties, and farmers were driving their stock to higher ground. 

During the night of the 24th-25th the rainfall was the heaviest of the flood period. Its 
effect is apparent in the river stages recorded on the morning of the 25th. The White River 
was out of its banks throughout its entire course and rising rapidly. By 9 a. m. the river at 
Indianapolis had reached 19.6 feet, passing the mark of 19.5 feet set by the flood of 1904. At 
Muneie the high water of 1904 was exceeded by 1 foot; 400 dwellings were surrounded by 
water; city water service was cut off; all railway communication discontinued. One hundred 
families had been driven from their homes at Johnstown. At Anderson 1,000 persons had been 
routed from their homes by water, Martinsville was cut off by washouts on steam and electric 
lines and many houses were under water. At Tipton more than 100 families had moved to 
higher ground; the city was in darkness, the light and power plant being under water. At 
Petersburg the White River rose 8 feet during the night ; every house in the western part of 
the city was flooded; mines and factories all closed. In Indianapolis the West Washington 
and West New York Street lowlands were inundated, and from these flooded districts the 
rising waters were spreading into the bottom lands of West Indianapolis behind the levee ; by 
noon the fires under the boilers in the power houses of the Indianapolis Water Co. and the 
street railway company were drowned, leaving the city without water for fire protection and 
without street-car service. Steam roads and trolley lines throughout the flood district were 
tied up by washouts and unsafe or down bridges ; towns were isolated ; the wildest rumors were 
rife of entire communities being swept away by constantly rising waters. The local office 
of the Weather Bureau issued the following warning : 

Below Indianapolis the river will rise rapidly and the public should prepare for higher stages than have 
been experienced for many years. Every precaution should be taken by those living along the lower course 
of the river, as the rise will be unusually rapid and will reach a point several feet above the danger line. 

The crest of the flood passed Indianapolis on the morning of the 26th. The Washington 
Street highway bridge, Indianapolis & Vincennes Railroad Bridge, Vandalia Railroad Bridge, 
and a private bridge belonging to Kingan & Co., no longer able to withstand the force of the 
water and the pounding of debris hurled against them, gave way during the clay. The levee 
along the west bank of the river, built to protect West Indianapolis from high water, began to 
crumble and give way at 10 a. m., allowing a great volume of water to rush into that portion 
of the city, but, as the people had been amply warned of the weakened condition of the levee 
and the probability of a break, none was taken unaware and few lives were lost, although 
many who refused to give heed to the warnings had to be rescued from the windows and roofs 
of houses and from trees and telephone poles. Many weird and pitiful stories are told of the 
suffering and privations of those who were obliged to spend the night perched on a roof or in 
trees. 



74 THE FLOODS OF 1913. 

Reports during the day gave the assurance that on the upper reaches of the river the flood 
was subsiding rapidly, though along the lower courses the river continued to rise, the crest not 
passing at Elliston until the 27th. 

The morning of the 28th broke bright and clear; no more rain fell during the rest of the 
month and the flood waters on the White River watershed continued to recede, rapidly in the 
upper courses of the streams and more slowly along the lower reaches. 

LOSS OF LIFE AND PROPERTY WHITE RIVER WATERSHED. 

When the high stages and rapid rise of streams are considered the loss of life is surprisingly 
small; only 12 deaths, 5 of which occurred in Indianapolis, being due directly to the flood. 

The property loss, while heavy and beyond question the greatest flood loss ever experienced 
in this district, is not nearly so great as at first estimated. The heaviest loss was suffered in 
the city of Indianapolis, where an area of approximately C square miles was inundated. This 
area comprised park, residence, and factory districts. Four thousand families were driven 
from their homes bj the water, and a majority of these suffered practically a total loss of their 
household goods. 

Crop losses for the district would have been much larger had the flood occurred during the 
crop season; as it was the damage was confined to winter wheat and hay crops. Great damage 
resulted from erosion and the deposit of sand and gravel over areas that had been fertile farm 
soil. In some localities whole farms were covered with sand to such a depth as to render the 
land worthless for farm purposes for many years; in others the top soil was washed away to 
such an extent that the land is now unfit for farming. Damage of this character occurred 
chiefly along the lower courses of the rivers. 

The values given in the following table have been secured from reliable sources and are 
believed to be a fair and conservative estimate of the losses sustained over the White River 
watershed : 

Money value of property destroyed or damaged, including public highways and bridges, also the 

cost of cleaning up basements and putting machinery and equipment in serviceable condition $1. 659, 855 

Money value of crops destroyed, or amount of damage thereto 262, 500 

Money value of live stock or other farm products lost 51, 150 

Money value of losses occasioned by enforced suspension of business through flood, including wages 

of employees 622. 600 

Loss in the State to railroads and trolley lines 4,807,555 

Loss in Wabash Valley 10.000.000 

Total loss in Indiana, all classes 20.103.660 



FLOOD ON THE WABASH RIVER IN MARCH, 1913. 



By W. R. Cade, Observer, United States Weather Bureau. 



On Sunday morning, March 23, river-gage readings from the different stations on the 
Wabash River indicated a stage of water somewhat below the normal height for the time of 
year. Bluffton reported a stage of 2.5 feet; Logansport, 3.8; Attica, 6.8, and at Terre Haute 
the stage was 7 feet. The river was falling at all points, and no one entertained the idea that 
within three or four days the greatest flood on record in the Wabash Valley would have reached 
its climax. A heavy rain fell on March 23 over the entire district drained by the river, and at 
7 o'clock the following morning 3.80 inches was measured at Bluffton, 2.82 at Logansport, 2.80 
at Attica, and 0.99 inch at Terre Haute. This heavy rain caused the river to rise at a record- 
breaking rate, and at 7 o'clock Monday morning, March 24, the river had risen to a stage of 
12.3 at Bluffton, 12.1 at Logansport, 15.9 at Attica, and 14.5 feet at Terre Haute. This meant 
a river above flood stage (12 feet) at Bluffton, Logansport, and Attica, and rapidly approaching 
the flood stage of 16 feet at Terre Haute. 

The heavy rain continued and the amounts on Tuesday were equally as large as those of 
Monday. The amount of rain over the valley during this period ranged from 2 to 3 inches, 
and the river continued to rise very rapidly. At 7 a. m. Tuesday, March 25, the following river 
stages were observed: Bluffton, 17. 5-; Attica, 24.6; Terre Haute, 19.5. During the afternoon 
of the preceding clay (Mar. 24) the bridge on which the river gage at Logansport was installed 
was washed away, so that further readings from this point were impossible. 

The next day, Wednesday, March 26, the rain was only slight, but the river continued 
rising. The crest of the flood passed Bluffton and Logansport on this date, Bluffton recording 
a stage of 20 feet and the Logansport's observer being unable to make a reading. 1 

During the 24 hours ending March 27 about half an inch of rain fell over the valley, which, 
fortunately, was the last measurable quantity of precipitation until March 31. On the 27th 
the crest of the flood reached Attica, a stage of 33.4 feet being reached, which was 21.4 feet 
above the flood stage, the highest water ever recorded. The highest water at Terre Haute was 
also recorded on this date. At 10 a. m. a stage of 31.3 feet was reached, and the water remained 
at this point for 4 hours, at the end of which time it began to fall slowly. 

When the river at Terre Haute reached a stage of 19 feet the residents of a little settlement 
of about 400 inhabitants, known as Taylorville, on the west side of the river, began to leave 
their homes. Within a day after they left their houses the entire place was submerged. The 
people living in Taylorville and in West Terre Haute were advised through the newspapers by 
the local office to prepare for a high river which was fast approaching the city. West Terre 
Haute, also on the west side of the river, suffered considerable damage during the flood. For 
the first time in the history of this town, which has a population of 4,000, the water covered 
its main street, and was several inches high on the floors of the stores and other business houses 
on this street. 

1 Later it was determined that the river reached a stage of 22.5 feet. 

75 



76 . THE FLOODS OF 1913. 

By Tuesday morning, March 25, the water had driven the Taylorville people from their 
houses, and the next day the lower parts of West Terre Haute were inundated'. It was on this 
day, March 26, that the river began its work of destruction on the Terre Haute side. The 
water, gas, electric, and other plants, including four large distilleries, along the river fought 
the rising water with sandbags, pumps, and other means. On the evening of March 26 a levee 
north of Terre Haute broke and a large section of the north end of town was covered with a 
foot or two of water. 

The river stage on Thursday morning, March 27, was 31.2 feet, and the water was rising 
slowly. A few hours after this reading the crest of the flood was reached, 31.3 feet. The water 
remained at this height for several hours and then began to slowly recede. At 8 a. m., this date, 
the city gas plant shut down, and the electric plant discontinued operations about 45 minutes 
later. The water had been bravely fought at 'these places, but when the river reached 31 feet 
it was necessary to abandon further efforts in this direction. 

Of course, with the shutting down of the electric plant, the street-car service of the city 
was stopped, and on the night of March 27 the city was wrapped in darkness. Very few 
business houses generated their own electricity, and business on this night was conducted in 
candle and lamp light. People walking along the main street of the city carried lanterns, and 
the streets in the resident parts of town were illuminated only by the faint beams from a candle 
or a lamp coming through an open window. The city was without electricity until early in 
the morning of the 28th, and gas was not supplied until nearly 24 hours later. 

After the water reached the 27-foot stage the city was practically cut off from the rest of 
the country. Railroad bridges were rendered unserviceable, and the telegraph companies 
would not promise to get a message off in any direction. The new bridge connecting Terre 
Haute with the town of West Terre Haute, on the other side of the river, was closed to traffic 
for two days, March 28 and 29. However, this bridge sustained only slight damage. On the 
west side of the river, the Vandalia and Big Four tracks were under water for a distance of 
a mile or more, and considerable damage was done railroad property in this place. 

Fortunately, the city was not afflicted as were some other places on the river in having the 
water cut off. While the water company could not keep its water up to the standard, it did not 
find it necessary to close clown. 

The damage, inconvenience, and suffering resulting ' from the flood at Terre Haute, was 
experienced at all other places along the river. At Logansport, La Fayette, and Clinton bridges 
were washed away. At most places the water, gas, and electric plants were stopped, which, 
coupled with the other afflictions that came upon the people as a result of the flood, made the 
situation most acute. 

At Logansport the bridge on which the Weather Bureau gage was installed was washed 
away, and the water rose to such a high point in this town that most of it was under water for a 
while. The rain gage was carried away, and also some of the records in the possession of the 
river observer. 

The water, gas, and electric plants of La Fayette had to discontinue operations ; the traffic 
on all steam and electric lines entering the city was suspended ; the heating plant was closed ; 
the fuel supply became exhausted, and one bridge was washed away and another condemned. 

Peru was the most demoralized of any town in the Wabash Valley. At that point three 
persons were drowned, and the suffering was most intense. All public buildings were turned 
into commissaries or dormitories. The water rose so rapidly and high that it caught people 
in their places of business and elsewhere. The people in the town of Peru, and elsewhere in 
the valley, expected the water to reach only the usual flood stage, and consequently the flood 
proved disastrous to many who thought they were safe. 

It is estimated that the property loss in the Wabash Valley will aggregate $10,000,000. 



A REPORT ON THE FLOOD IN THE ILLINOIS RIVER IN MARCH, APRIL, 

AND MAY, 1913. 



By Montrose W. Hayes. Local Forecaster. 



A joint resolution of the Illinois Legislature, passed in May, 1889, relative to constructing 
a waterway from Lake Michigan to the Mississippi River, refers to the Illinois River in the 
following language : 

The Illinois River from La Salle to Grafton is the remnant of an ancient stream bed, bordered by wide 
and low bottom land, much cut up by lake, bayou, and marsh: and an alluvial stream of small low-water 
volume and sluggish current, with a declivity of only 2G feet in 225 miles, a declivity so small as to require 
a large volume of water to maintain an effective channel ; a stream which in its natural condition is able 
to maintain but a small depth through the deposits with which the tributaries constantly tend to choke 
the channel, a tendency ever increasing with the inhabitation of the watershed and the cultivation and 
reclamation of lands. 

It is quite probable that this alluvial stretch of the Illinois, which really begins at Utica, 
6 miles above La Salle and 230 miles above Grafton, has less slope than any other stream in 
the upper three-fourths of the Mississippi River drainage area. It naturally follows, on account 
of the slight slope, that the run-off is slow and the gradual rises and falls cause the stream to 
remain out of its banks or bank full after other watercourses flooded by the samfe storms have 
fallen to normal stages. 

The beginning of spring almost always finds the Illinois nearly or quite bank full, even 
though the winter precipitation may have been less than the normal. This, is due to the effect 
of cold weather on the water flow. The river freezes over, friction caused by the ice covering- 
decreases the discharge, and the fall in the stages is not proportional to the decrease in the 
tributary increment that follows the freezing of the small streams. When the ice begins to 
move the small slope causes it to gorge and still further decrease the discharge, and the tributary 
water released as the weather grows milder runs into a trunk stream that is almost or altogether 
full. It is evident, therefore, that the spring precipitation does not have to be especially heavy 
or long continued to give flood stages. 

A study of gage readings, in connection with precipitation charts, shows that the stretch 
of river below the mouth of the Sangamon is, at high stages, influenced materially by the stage 
of the Mississippi at Grafton. (The Sangamon flows into the Illinois 98 miles above Grafton.) 
This influence, of course, is felt throughout the extent of the alluvial stretch, but above the 
mouth of the Sangamon it is not sufficient^ pronounced to show itself in a casual investigation 
of the gage readings and precipitation charts. 

In 1913 the precipitation over the Illinois watershed was about normal in January and 
somewhat less than normal in February. When March opened the river was bank full through 
most of the alluvial stretch, as is usual for the season, and frequent moderate rains caused a 
gradual rise until the latter third of the month, when the passage of a phenomenal procession 
of storms began. These storms were not productive of as much precipitation as they were in 
Indiana and Ohio and in that portion of Illinois drained by streams that enter the Mississippi 

77 



78 THE FLOODS OF 1913. 

south of Grafton, but they were heavy enough to cause a rise to begin simultaneously in all 
stretches of the river. Eain fell on March 20 and 21, and from the 23d to 27th, inclusive. The 
combined amounts for the two periods ranged between 2 and 4 inches over the watershed above 
the mouth of the Sangamon and between 4 and 6 inches over the remaining or lower part of 
the watershed. The precipitation in the country, receiving between 2 and -t inches, was note- 
worthy more because it was so general and uniform than because it was excessive. It was not 
nearly so heavy as the rainfall of September, 1911. 

At La Salle the rise was comparatively rapid, as this station is only 6 miles below the upper 
extremity of the alluvial stretch, and the crest, 28 feet, occurred on March 29; but with the 
entire river from Utica to the mouth overflowing the fall naturally was much slower than the 
rise, and the stream was not within its banks until April 18. 

At Peoria, 62 miles below La Salle and 162 milas above the mouth of the river, the highest 
stage was reached on March 30 and prevailed until April 2. Peoria is at the foot of Peoria 
Lake, which is a succession of three expanses of a total length of 17 miles, in which the river 
spreads in places to a width of a mile and a half when the stages are normal. This lake, as it 
is termed, has a reservoir effect that gives, even for the Illinois River, very gradual fluctuations 
in stage at Peoria. There was practically a continuous fall after the maximum, which was 22.3 
feet, occurred, but it was the latter part of April before the stream was in its banks. 

Beardstown is 135 miles below La Salle and 89 miles above the mouth of the river. At 
that station the highest stage was 21.8 feet which prevailed on April 5 and 6. The crest of the 
flood at Beardstown did not have the same form the crest at Peoria had, as it was materially 
changed by the volume of water contributed by the Sangamon River, which flows into the 
Illinois 9 miles above Beardstown. After the passage of the crest there w T ere heavy rains over 
the lower part of the Illinois watershed, and the river was bank full at Beardstown until 
May 13. 

This flood was an unusualty high one, but none of the stages was unprecedented. Most 
of the conditions were favorable for a record-breaking high-water level, but one very important 
requisite was missing; that was a high stage in the Mississippi at the mouth of the Illinois. 
All of the floods that were greater than this one have been coincident with a flood in the Mis- 
sissippi at Grafton, but at no time in the spring of 1913 was the Mississippi stage more than 
what may be termed bank full at Grafton, and on this account there was a good current in the 
Illinois even when the water was highest. This comparatively rapid discharge, it is believed, 
prevented the river from reaching a record-breaking height, at least in its lower reaches, where 
the influence of the Mississippi stages is most pronounced. 

Levees broke at Hennepin, 208 miles above the mouth, at Meredosia, 71 miles above the 
mouth, and at Naples, 65 miles above the mouth. At Beardstown the levee protecting the 
Combes addition to the town broke. All of these breaks caused considerable property loss, 
and inconvenience and suffering to those who were compelled to abandon their homes. At 
Meredosia the Wabash Railroad crosses the river, and the railroad approaches or embankments 
were damaged to such an extent that traffic had to be suspended. 

Along the entire alluvial stretch of the river the water was high enough to damage road- 
ways of all kinds, bridges, buildings, etc., but the greatest damage was on agricultural lands. 
The tributaries were carrying a large volume of water which could not be discharged and 
cross levees were washed out or damaged. Considerable corn had been left in the field and 
most of this was lost. Some farm buildings and machinery were damaged or entirely ruined, 
and some wheat was lost by being covered by water; however, from the reports received, it is 
not thought that the loss to growing crops was large. 

The slow fall of the flood water caused considerable anxiety to farmers; it was feared 
that the ground could not be put in condition soon enough to enable a crop of corn to be put, 
in, and in most sections along the river the planting was, necessarily, late, but not too late 
for the crop to give a good yield, provided favorable weather prevails during the remainder 
of the season. 



FLOOD IN THE ILLINOIS EIVEB IN MARCH, APRIL, AND MAY, 1913. 79 

There was no direct loss of life in the Mood: that is to say, the Weather Bureau corre- 
spondents along the river reported none, and no newspaper accounts of any were seen. Many 
families were driven from their homes by breaks in the levees, or overflows, or through fear 
of the water coming in, and it is quite probable that there were deaths from exposure, which 
should be termed deaths caused by the flood, but there appears to be no way of obtaining sta- 
tistics of this. In fact, all the statistics of losses are quite unsatisfactory, and in reality mean 
very little. To obtain figures that could be considered authentic, as well as complete, would 
be an expensive and tedious task, probably out of proportion to the value of the information. 
And when the losses in Illinois are compared to those in Indiana and Ohio they become so 
insignificant they are of minor importance. 



THE FLOOD IN THE OHIO, IN THE LOUISVILLE DISTRICT. 



By F. J. Walz, Professor of Meteorology. 



The flood was caused by excessive rains over the entire drainage basin during the period 
March 23-27, 1913. The rainfall was much greater over the drainage areas of the northern 
tributaries of the Ohio than over the southern tributaries of that stream, the reverse of the 
distribution that produced the floods in January last. The flood waters, therefore, in the 
smaller streams and in the main tributaries flowing through Kentucky, except over the Ken- 
tucky River section, were not so high or so destructive as they were in January. The heaviest 
rains in Kentucky, 6 to 7 inches, fell over the upper watershed of the Kentucky River, most 
of it in about 48 hours, hence the greater flood in that stream in comparison with other streams 
in the State. Also between 6 and 7 inches of rain fell in about the same period of time over 
the upper watershed of the Cumberland and Barren Rivers, causing floods of large propor- 
tions in sections of those streams. In other than the above-mentioned sections, the flood 
damage, while large, was not extraordinarj' except from the overflow of the Ohio River itself. 
Away from the Ohio, while damage to property runs high yet there was comparatively little 
suffering or distress and losses were mostly from interruption to traffic and business. 

(The daily amounts of precipitation that occurred over Kentucky during the period 
Mar. 23-27, 1913, will be found in Table 2, p. 22.) 

At Pikeville, Ky., on the Big Sandy River, the crest of the flood was reached at 2 p. m., 
March 27, 39 feet. At Louisa, Ky., on the same stream the crest Avas reached at 12 noon 
March 28, 44.8 feet. While this stage is 15.3 feet higher than the stage reached in January, 
the location here is such that stages in the Big Sandy above 20 feet are due almost entirely to 
backwater from the Ohio River. 

At Falmouth, Ky., on the Licking River., a crest of 34.1 feet Avas reached at 1.30 p. m. 
March 27. This was 0.3 foot below the reading reached January 12, 1913. 

At Rumsey, Ky. (Lock No. 2), on the Green River, the flood in January reached a crest 
of 35.5 feet, while the highest reached in this flood was 31.2 feet, April 5. 

At Burnside, Ky., on the upper Cumberland RiA^er, the crest reached in January and that 
in March were exactly the same, 61.5 feet. The high flood stage reached at Burnside in the 
March flood Avas due to the fact that the northern headAvater tributaries of the Cumberland 
drain an area visited by the hea\*y rains Avhich also extended over the upper tributaries of the 
Kentucky River. 

KENTUCKY RIATEK. 

The Kentucky River was above flood stage (30 feet) at Beattyville on two days, the 27th 
and 28th, reaching a crest at 2 a. m. of the 28th of 39.9 feet which is 7.9 feet above the January 
stage. The estimated damage in Beattyville and vicinity is only about $5,000 from all causes. 
The money value of property saved by moving and protection is about $10,000. 

At High Bridge the Kentucky River Avas above flood stage (27 feet) from the 27th to 
the 31st, inclusive, and reached a crest of 34.6 feet on the 27th. This is 1.6 feet abo\ v e the 
crest reached in January. The damage was inconsiderable, in fact about $1,000, mostly from 
14284°— 13 6 81 



82 THE FLOODS OF 1913. 

loss of saw logs which were carried away. The flood stage at High Bridge has since been 
raised to 30 feet. 

At Frankfort the Kentucky River was above flood stage from the 27th to 31st, inclusive, 
and reached a crest of 3S.3 feet at 5 a. m. on the 28th, which was 4.5 feet above the crest of 
January 12, 1913. The highest water known at Frankfort, 44.5 feet, occurred in February. 
1878. The waterworks pumps at Frankfort were again put out of commission, and there was 
some inconvenience and suffering from the water famine; also there was considerable damage 
from landslides on steep banks in that vicinity. Train service was interrupted for a few days, 
otherwise the damage there was inconsiderable and not any greater than in January last. 
Flood warnings were issued to the river observer at Frankfort, beo-innino- with the 26th, and 
from there widety distributed by telephone, and also through the Louisville newspapers which 
circulate largely in all parts of the State. These warnings enabled a number of families to 
leave their homes and remove their belongings before the water reached them. 

OHIO RIVER. 

At Madison, Ind., the river was above flood stage — 46 feet — from the afternoon of March 26 
to the early morning of April 8. The crest of the flood reached a stage of 62.8 feet, which is 
the highest on record and was 1 foot above the record of February, 1884. This was the only 
station in the Louisville district on the Ohio River where the stage of 1884 was exceeded and 
was no doubt due to the coming together of the extraordinary flood waters of the Great Miami 
and those of the Kentucky. At Cincinnati, Ohio, a crest of 70 feet was reached, which is 1.1 
feet below the crest of the 1884 flood. 

At Louisville. Ky., a crest of 44.9 feet was reached about 2 a. m. of April 2, which is 1.8 
feet below the 1884 flood crest. Here the Ohio River was above flood stage from the afternoon 
of March 26 to the early morning of April 9. For the second time this year the river 
completely submerged Shippingport, a small section of the city lying between the river and 
the Louisville and Portland Canal and containing 78 families, also a section of the city below 
what is Itnown as the " cut-off," where some 1.700 families were rendered homeless. The flood 
water spread over quite a large area bordering on the river, including a part of a section 
known as Parkland and sections along Bear Grass Creek. The backwater up this creek 
flooded several of the principal streets, inundating bridges, which caused a disarrangement 
and partial suspension of street-car and other traffic. Also, there were a number of factories 
and mills of various kinds that were flooded and had to suspend business for a week or ten 
days at considerable loss. A large whisky warehouse containing 3.690 barrels of whisky 
collapsed on account of being undermined by high water, and the barrels of whisky were 
plunged into the river. Most of them were finally recovered. 

Railroad and mail services were badly hampered. All roads from Louisville to Chicago, 
except the Chicago, Indianapolis & Louisville Railroad (Monon), had to suspend business, 
and all train service ceased for several days on lines leading north and east except the 
Louisville & Xashville to Cincinnati. Roads from Louisville to St. Louis also suspended 
train service for several days. 

In Louisville and vicinity it is estimated that 2.000 persons were thrown out of employment 
on account of the flooding of factories. Damage to factories and business houses in the 
flooded district together with the loss in wages and from suspension of business is estimated 
at half a million dollars. River navigation was practically suspended on account of the 
high water making it impossible for the larger boats to pass under the bridges. The Union 
Railroad Station at the foot of Seventh Street was abandoned March 30, the water being over 
the tracks and in the lower story of the building, and all the railroad yards on the river front 
were for the most part under water for about a week. 

The city authorities were advised well in advance of the overflow and inundation of the 
two sections of the city where danger was greatest, namely. Shippingport and the Point 



THE FLOOD IN THE OHIO, IN THE LOUISVILLE DISTRICT. 83 

(or section below* the "cut-off"). As the flood waters were coming on so rapidly the mayor 
was advised by personal call over the telephone to get the people out of those sections without 
any delay or loss of time. With the aid of the police and fire departments all the families 
and their household goods were removed to places of safety before the flood waters enveloped 
their homes. 

One of the most critical and dangerous situations developed by the flood was the condition 
at Jeffersonville, Ind., across the Ohio River from Louisville. Ky. The city is protected by a 
levee in front and by an embankment or fill under the tracks of the Pennsylvania Railroad at 
the lowest point between the city and New Albany. The levee was not high enough to resist 
the stage of 45 feet and was strengthened at many places with bags of sand, cement, dirt, etc.; 
while across several of the streets where the levee was low, timber barricades were built and 
filled in with dirt to bring the height above the expected crest stage. But the hardest struggle 
was necessary at the Pennsylvania fill. As the water rose at this point the embankment began 
to slip away and leak. Electric lights were strung along the fill and a guard established; 
also about 200 of the convicts at the Indiana reformatory, a few hundred feet away, were 
pressed into service, and with their help about 170,000 sacks of sand and cement, great 
quantities of dirt and other material were used to reenforce and strengthen the weak places, 
the fight going on both day and night until the crest was passed and the water receding. About 
100 tarpaulins were loaned by the Jeffersonville depot of the Quartermaster's Department and 
were spread on the side of the fill next the river, materially assisting in preventing the water 
from cutting through the embankment. If this fill had broken, nearly every business house 
in the city and probably 2,400 homes would have been flooded, with enormous damage. On 
account of the close proximitjr of the local office of the Weather Bureau the city officials and 
citizens were able to obtain warnings at all times and made their desperate fight against the 
rising waters on the basis of the information furnished. The loss to property located outside 
the district protected by the levee is estimated at about $40,000, and as that part of the city is 
relatively small it is estimated that the value of property' saved by the work on the levee and 
fill is at least $1,000,000. 

Flood information and warnings were issued from the Louisville office from time to time, 
beginning March 26, 1913, of which the first and last only are given below: 

March 26 (10 a. m.). — Excessive rainfalls, ranging in amount from 2 to 5 inches and more, have fallen 
during the past 24 hours over the entire watershed. The river has risen 21 feet since yesterday morning at 
Cincinnati, and is now above flood stage at that point. It. had risen 16 feet at Madison and nearly 11 feet at 
Louisville at 7 a. m.. and 2 feet at Louisvile from 7 to 10 a. m. It is now rising at the rate of nearly one-half 
foot per hour. It will pass the flood stage early to-night and continue rising rapidly during several days. 

Special bulletin (9 p. m.). — At 7 p. m. the river stage at Cincinnati was 54. S feet and rising at the rate 
of three-tenths of a foot per hour. It was rising steadily above Cincinnati, but in the stretch from Catletts- 
burg to Maysville it was 15 to 20 feet below flood stage, while at Pittsburgh it was 3 feet above flood stage. 
At 9 p. m. the stage at Louisville was 30 feet, and the river was rising at the rate of five-tenths of a foot per 
hour. A stage of near 33 feet or more is expected by 7 or S o'clock Thursday morning. It will continue rising 
during Thursday, but the rate of rise will probably decrease somewhat, and from present indications the crest 
will be between 35 and 36 feet at Louisville. 

April 7, 1913. — The river this morning has passed below the flood stage at Maysville and was only 0.5 foot 
above that stage at Cincinnati, where the fall since Saturday morning has been nearly 13 feet. The decliue 
in this section will from now on be rapid. 

The bulletins were given wide dissemination through the local press and over the telephone. 
In addition a bulletin giving the river stages at the various stations reporting, at stated times, 
was issued daily in postal-card form and mailed to persons interested. This bulletin also con- 
tained a statement as to expected heights as far as they could be anticipated. Thousands of 
inquiries were received over the telephone, many of them long distance. In fact, at points from 
the mouth of the Kentucky Eiver to Tell City, Ind., a distance of nearly 200 miles, the people 
were kept directly informed and warned through the medium of the telephone. 

It is needless to say that the services of the Weather Bureau proved of inestimable value 
to the flooded sections and those threatened by flood. Many commendations of the value and 
accuracy of the warnings and forecasts have been received. 



OHIO FLOOD— EVANSVILLE, IND., DISTRICT. 



By Ax. Brand, Local Forecaster. 



Continued heavy rains over most of the Ohio and its tributary valleys during March 24, 25, 
26, and 27 started a rapid rise along this stretch of the Ohio River on March 25 from stages 
ranging between 24 and 27 feet. Within 48 hours after the rise began flood stages were reached 
and passed at Evansville, Henderson, and Mount Vernon, and devastating, record-breaking 
flood conditions prevailed in most of the Evansville River district from March 26 to April 18, 
inclusive. 

At Evansville, Ind., the rise began on March 25, from a stage of 26 feet. The rise was 
especially rapid during the first 48 hours, amounting to 10.6 feet ; river reached and passed the 
flood stage, 35 feet, sometime during the hour ending at midnight of March 26. The greatest 
24-hour rise, 6-5 feet, occurred during the period ending at 7 a. m., March 27. From 7 a. m., 
March 27, to 7 a. m., April 1, inclusive, the 24-hour rises diminished gradually from 3.8 feet to 
1 foot. After 7 a. m. of April 1 the 24-hour rises were less than 1 foot until about 9 a. m. of 
April 3, when the river came to a stand with the gage indicating a stage of 47.9 feet; river 
remained stationary at a stage of 47.9 feet until shortly after 12 noon of April 3, when a second 
creeping rise set in as a result of a heavy rain which had set in over this section of the Ohio 
and Green River Valleys, combined with a fresh to brisk wind blowing directly onto the gage. 
The second rise terminated shortly after 12 noon of April 5, with a crest stage of 48.3-}- feet. 
This crest stage of 48.3-(- feet is the highest stage ever recorded at Evansville in the history of 
the station. River remained stationary at 48.3+ feet until about 6 p. m. of April 5, when it began 
to fall slowly ; the slow fall, amounting to less than 1 foot in 24 hours, continued until 7 a. m. 
of April 10, after which the 24-hour rate of fall increased slowly, but gradually, from 1.4 to 
2.5 feet ; river was falling at the latter rate when it dropped below the flood stage, 35 feet, at 
about midnight of April 15. Since that time the river has continued falling at a rate averaging 
slightly over 1.5 feet in each 24 hours. 

At Henderson, Ky., where the rate of rise and fall was again very similar to that experi- 
enced at Evansville (the greatest 24-hour rise, 6.5 feet, being the same and occurring on the 
same date as that at Evansville), the rise started on March 25 from a stage of 24 feet; river 
reached and passed the flood stage, 33 feet, shortly before daybreak of March 27; river came 
to a stand during the afternoon of April 5, with a crest stage of 47.9 feet. The crest at Hender- 
son was, it is believed, materially raised by backwater from the Wabash. River remained 
stationary at Henderson with a stage of 47.9 feet until sometime during the night of April 5 
and 6, when it began to fall slowly; river fell below the flood stage, 33 feet, at 11.30 a. m., 
April 16. 

At Mount Vernon, Ind., the rise began on March 25, from a stage of 26.3 feet. During the 
first 24 hours ending at 7 a. m., March 26, the rise amounted to 2.7 feet, while the greatest 24- 
hour rise, 5.3 feet, occurred during those ending at 7 a. m., March 27. River reached and passed 
the flood stage, 35 feet, at 11 a. m., March 27. From 7 a. m., March 27, to 7 a. m., March 30, 
inclusive, the 24-hour rises gradually diminished from 3.9 to 2.7 feet. As a result of the 
destructive flood wave which began surging into the Ohio from the Wabash during the night of 

85 



86 THE FLOODS OF 1913. 

March 29 and 30. the rate of rise at Mount Vernon again increased quite suddenly during the 
morning of March 30, due to backwater, and amounted to 3.5 feet during the 24 hours ending 
at 7 a. m. of March 31 ; after this time the rate of rise diminished materially, the river coming 
to a stand at 9 a. m. of April 5, with a crest stage of 52.9 feet, scant. River remained stationary 
at a stage of 52.9 feet, scant, until sometime during the night of April 6, when it began to fall 
-lowly ; the 24-hour rate of fall amounted to less than 1 foot until after 7 a. m., April 10, when it 
increased gradually from 1.7 to 3.5 feet; river was falling at the latter rate when it dropped 
below the flood stage, 33 feet, at 4 a. m., April 18. Since falling below the flood stage, the decline 
has continued steadily, mostly at a decreasing rate. 

ACTION BEFORE AND DURING FLOOD. 

The first reference to the rapid rise expected along this stretch of the river was made in 
the regular river forecast published on March 25, which read as follows: 

The Ohio. — As a result of the heavy rains during the past 48 hours, a rise will set in at Evansville, 
Henderson, and Mount Vernon this afternoon or to-night. The rate of rise will increase rapidly during the 
next 48 hours. Sufficient water now in sight to raise the present stages in the lower part of the Evansville 
district 3 or 4 feet by Wednesday night or Thursday morning. 

At 8.40 a. m. on March 26 (Wednesday), the following flood warning was issued by the 
official in charge of the local office and disseminated as widely as possible throughout the 
Evansville River district by means of the local press, river bulletins, telegraph, and telephone : 

Flood warning. — Stages of 35. 33, and 35 feet will he reached and passed at Evansville, Henderson, and 
Mount Vernon, respectively, Wednesday night or Thursday morning. Notify interests. 

From time to time additional warnings Avere issued as conditions of the weather demanded. 
Only the initial and final warnings are here given. The latter, under date of April 4, 1913, 
follows: 

April J/. — The Ohio River continues falling at all upriver points. The rain during the past 24 hours, 
combined with backwater and brisk wind, has caused additional slight rises on all gages in the lower part of 
the district. While the river is now on a stand at Evansville, additional rises of a tenth or so may occur at 
these points due to the rain, backwater, and wind. At Mount Vernon it will come to a stand to-night or 
Saturday morning, with a stage of about 53 feet. 

In addition to the flood warnings and river forecasts published and distributed by every 
possible means from day to clay, the official in charge of the local office furnished by telephone 
and telegraph advisory and timely information to all applicants from outlying districts and 
neighboring river towns; when the water began to encroach upon Water Street he also per- 
sonally called on the merchants on that street and advised them to make use of splashboards 
and sand bags in order to keep the water from getting into their basements and first floors. 

The March-April, 1913, flood was, at least in the lower part of the Evansville River district, 
the greatest and also the most devastating flood in the history of this station, as well as in the 
memory of the oldest inhabitant. All of the lowlands and most of the river towns and cities in 
the district contributed toward making up the enormous losses experienced in the Ohio Valley 
from Louisville, Ky.j down to the mouth of the "Wabash River. The flood water rendered 
homeless thousands of people and made prisoners of many others in their homes for clays, in 
some instances without food. Thousands of buildings, consisting of houses, barns, outbuildings, 
tool sheds, corncribs, etc., with most of their contents, were completely washed away, while a 
great many that were left were thrown or moved from their foundations or otherwise dis- 
arranged ; hundreds of thousands of bushels of corn and immense quantities of cured hay and 
other foodstuffs were lost, and large tracts of growing wheat, clover, timothy, etc., were com- 
pletely destroyed. Some, in fact most all, of the rural districts were deprived for days of all 
means of outside communication, while some of the smaller river towns were without mails for 
a week or 10 days. Enterprise. Tell City, and Cannelton, Ind., and Hawesville and Uniontown, 
Kv.. were, it is believed, the heaviest sufferers among the towns affected by the flood in this 
district. 



OHIO FLOOD EVANSVILLE, IND., DISTRICT. 87 

At Enterprise, Ind., a small village, all the inhabitants. 30 families, or about 120 people, 
were driven from their homes, as every house was more or less flooded. 

At Tell City, Ind., a thriving and prosperous manufacturing town, with a population of 
about 4,000, all of the factory district as well as a large portion of the residence section was 
flooded, and about 300 families were compelled to abandon their homes. 

At Cannelton. Ind., and Hawesville. Ky.. two small towns on nearly opposite sides of the 
river, each with a population of about 1.000. the greater portion of each town was flooded, and 
about 75 families of each place were required to seek temporary homes elsewhere. 

At Uniontown, Ky., 13 miles below Mount Vernon, Ind., and about G miles above the 
mouth of the Wabash River, with a population of about 2.000, where the suffering, as a direct 
result of the flood, is believed to have been infinitely worse than at any other point along the 
Ohio, not excluding Shawneetown, 111., every house in the city was flooded to such an extent 
as to make it absolutely uninhabitable, and all residents were compelled to leave their homes 
and seek shelter at the fair grounds, back on the hills. A member of a relief party sent out 
from this city (Evansville) stated that the actual suffering at Uniontown during the first three 
or four days after the residents of that place were compelled to leave their homes was inde- 
scribable; rich and poor alike Avere compelled by necessity to huddle together in stalls and in 
such other improvised shelters as were offered by the flimsy and insecure structures at the 
fair grounds, without the necessary food, bedding, or clothing. 

The publisher of the Telegram at Uniontown, Ky., states that the losses to city property 
at Uniontown, Ky., will amount to at least $200,000; that there was a loss of about 60,000 
bushels of wheat and about 75,000 bushels of corn in that immediate vicinity, and that the 
losses occasioned by enforced suspension of business, including loss of wages by employees, were 
estimated to amount to another $50,000. 

At the present market value of wheat and corn the losses at Uniontown, Ky., and vicinity 
would in the aggregate amount to fully $33G,500. The publisher of the Telegram also stated 
that — 

Possibly half of this loss would have been avoided if the people had heeded the warnings sent out by 
the Weather Bureau. Most of the people prepared for an 1S84 water only, and when the crest reached 27 
inches higher the greatest part of the damage was suffered. 

From information obtained from other sources it appears that most of the residents of 
Uniontown, Ky., notwithstanding the warnings sent out by the local office. AVeather Bureau. 
Evansville, Ind., believed that they would be able to escape flood water by scaffolding about 
1 inch higher than what would have been necessary to escape the last January water. As a 
result the upriver water, reenforced by the flood wave which began surging out of the Wabash 
during the night of March 29-30, caught them unprepared for the sudden rise which occurred 
at Uniontown on March 30 and 31. The sudden rise, clue largely to backwater from the Wabash, 
prevented the removal of the merchandise in the stores, as well as the furniture, etc., from the 
houses ; it also prevented rescaffolding of same to raise it above flood water. The losses resulting 
from the flood will be a serious blow to the merchants of Uniontown. 

The main cause of the destructive flood wave which swept things clean throughout the lower 
White and Wabash Valleys on March 28, 29. and 30 appears to have been the breaking away 
of about 1,400 feet of the Chicago & Eastern Illinois Railroad embankment, located in the 
White River Valley, between Decker and Hazelton, Inch, and about 18 miles above the mouth 
of the White River. This embankment, which is about 4 or 5 miles long, acts as an immense dam 
in times when the White River is in flood ; in consequence of this, when it broke late Friday 
afternoon or evening, March 28, a perfect wall of water was suddenly released and allowed to 
sweep everything before it in the lower White and Wabash Valleys. 

At Evansville, Ind., the flood caused but little damage but considerable inconvenience. 
All of the main portion of the city was high and dry, the only districts flooded being those 
along the eastern and southeastern edges of the city known as Oakdale, Crofton Place, Maple 
Place. Maple Grove, and Garden Acres ; some of the low-lying sections along Pigeon Creek, and 



88 THE FLOODS OF 1913. 

Water Street, between Vine and Locust Streets, were also flooded. The suburb of Oakdale, 
which is believed to be the lowest section of the city, is composed of 400 or 500 modest houses, 
averaging in value between $900 and $1,000, and begins to flood when the river passes the 
42-foot stage on the local gage. "When the water was at the maximum stage the flooded districts 
along the eastern and southeastern edges of the city, as well as those along Pigeon Creek, were 
flooded to a depth ranging between 2 and 61- feet, the latter depth being confined almost entirely 
to the lowest sections of Oakdale. At a stage of 46.5 feet the water begins encroaching upon 
Water Street adjoining the public levee. When the flood was at its highest all of the business 
portion of Water Street, between Division and Locust Streets, was flooded to a depth of about 
6 inches, the water reaching almost to the top of the curbing along the outer edge of the side- 
walks; some of the merchants along this street made use of splashboards and sandbags to 
prevent the water from being splashed into their first floors and cellars by the high wind. All 
railroads and some of the suburban electric lines entering this city, as well as the city street car 
lines in outlying sections, were seriously interfered with; all of the railroads, except those leav- 
ing and entering the city by way of the Louisville & Nashville Railroad Bridge, were com- 
pelled to suspend all through business to northern, eastern, and western points. This interrup- 
tion in railway traffic, which was general throughout most of the States bordering on the Ohio 
River, had a far-reaching effect, as it continued for about two weeks, and brought about an 
unusual sluggishness in all lines of trade in every city in this district ; it also interfered to a 
great extent with the prompt movement of mails throughout all of southern Indiana. 

The flood caused but little damage at Mount Vernon, Ind., and at Henderson. Ky., as both 
of these towns are located well above the flood line; they were, however, affected and incon- 
venienced by the interrupted mail and railway services. 

Local, State, and Federal aid in the way of tents, food, clothing, medicine, and feed for 
live stock was furnished to flood refugees at nearly all towns and cities in this district. 

Very little drift was left on the lands flooded, and most of the bottoms that were over- 
flowed were materially benefited by a deposit of silt in the nature of a rich loam, ranging between 
2 and 6 inches in depth. Spring plowing, it is believed, was not delayed by the flood. 



FLOOD IN THE MISSISSIPPI RIVER, APRIL, 1913, CAIRO, ILL., TO 

HELENA, ARK. 



By S. G. Emery, Local Forecaster. 



Following the rise in the Mississippi River that culminated on February 3, with a stage 
of 40.5 feet on the Memphis gage, the river fell rapidly, reaching 16.1 feet at New Madrid on 
February 26 and 17.3 feet at Memphis on February 27. Then, as a result of heavy rains in 
Tennessee and Kentucky and general though moderate rains over the northern watershed of 
the Ohio River during the last two days in February, the morning reports on March 1 showed 
rising stages in the Ohio River from Pittsburgh to Cairo and in the Mississippi from St. Louis 
to Memphis. In this district the total increase on river stages, due to the rains above men- 
tioned, was about 9 feet, the rise occupying the period March 1 to 10. This rise did not greatly 
affect the lower reaches of the Mississippi River, the increase at Vicksburg being only 1.4 feet, 
but inasmuch as they had not then regained normal conditions following the February rise it 
caused the abnormal stage to be sustained until the arrival of the southwest storm that swept 
over this section on March 26 and 27, attended by a heavy downpour of rain that caused an 
additional rise that continued until the Ohio flood waters began to enter the Delta country. In 
other words, the lower river was well filled before the real flood entered the head of the valley, 
a condition that usually attends all severe floods in the lower Mississippi. 

On March 15 the rise that had been in progress in the upper Mississippi and Missouri 
Rivers for some days began to be felt at Cairo, and, being augmented by a heav}^ downpour of 
rain an March 14, the rise was very rapid, as much as 15 feet being added to the Cairo gage 
in eight days. 

At Memphis the rise began on March 18, with a stage of 20.7 feet, and in the next eight 
days 11 feet was added to the gage reading. At Helena the stage increase in the first 11 days 
was 14 feet. Flood stage (35 feet) was reached at Memphis on March 30, the increase in 13 
days being 14.8 feet. The rise in the lower river up to this time was caused by a storm that 
passed northeastward over the Central Valley region on March 13 and 14, causing heavy rains 
and snows in the Missouri and upper Mississippi watersheds and general rains in the Ohio 
Valley from below Pittsburgh to Cairo. The principal flood waters from the Ohio River, those 
caused by the remarkable rainfalls in the States of Indiana and Ohio and which resulted in 
severely inundating several cities in those States, had not yet reached as far south as Memphis. 
After reaching flood stage at Memphis the rise was less rapid for a day or two, but when the 
first effects of the Ohio flood began to be felt the rate of rise increased to over a foot per day, 
and on April 7 reached 44.9 feet. At 8 a. m., April 8, the river stood three-tenths of a foot 
above the record for 1912, which was 45.3 feet. On the following clay, April 9, four-tenths of 
a foot had been added to the Memphis gage, and at 7 a. m., April 10, the gage reading was 46.5 
feet, with indications that a very slight fall had occurred between 5 a. m. and 7 a. m., estimated 
at less than a tenth of a foot. This stage, 46.5 feet, is 1.3 feet above the 1912 record and 9.4 feet 
above that of 1897, which up to that time was the greatest flood on record. A further rise 
at Memphis was prevented by a break in the levee at 5.30 p. m., March 9, at Golden Lake, Ark., 
38 miles by river above Memphis, and near the scene of the Wilson break in 1912. Also by a 
break in the levee near Graves Bayou, Ark., 23 miles below Memphis. The latter occurred at 
2 a. m., April 10, and was the direct cause of the fall shown on the Memphis gage of 1.8 feet 
on April 11. Following this decided drop, the next seven days showed a decline of less than 

SO 



90 THE FLOODS OF 1913. 

2 feet. At the time the breaks in the levee occurred the river at Memphis was rising at the 
rate of about one-half foot per day, and notwithstanding the fact that Cairo had reported a 
stationary or falling stage during the six preceding days the river continued rising at Cotton- 
wood Point. Mo.. 116 miles above Memphis, up to the 10th, and on the same day at Fulton, 
Tenn.. 57 miles above Memphis, showed a rise of one-tenth of a foot, and the next day, AjDril 
11. a fall of seven-tenths of a foot, which fall was clearly due to the break in the levee at Golden 
Lake. 20 miles below. From the fact that rising stages continued at all points south of Xew 
Madrid. Mo., until the levee gave way it is fair to assume that had no break occurred the river 
at Memphis would have continued rising at least two or three days longer and carried the 
stage to 47.5 feet. The conditions affecting the stage of the river at Cairo in 1913 were prac- 
tically the same as obtained in 1912; that is, the levees in that vicinity broke in the same places, 
and yet with only eight-tenths of a foot increase in stage height above 1912 at Cairo the gage 
at Memphis showed 1.2 feet more in 1913 than in 1912. This may be accounted for in part by 
the fact that only one break in the levee occurred below "Wilson, Ark., this year against two in 

1912. namely. St. Clair and Wyanoke. 

At Xew Madrid. Mo., the river reached flood stage. 34 feet, March 27, and the crest stage, 
44.5 feet. April 9. It was above flood stage 30 days, and above 40 feet 21 clays. In 1912 it 
was above flood stage 34 days and above 40 feet 23 days. 

At Memphis the river reached flood stage on March 30, and the crest stage, 46.5 feet, 
April 10. It was above flood stage 30 days, and above 40 feet 20 days. In 1912, it was above 
flood stage 56 days, and above 40 feet 23 clays. 

At Helena it reached flood stage on April 1. and the crest stage. 55.2 feet. April 21. It 
was above flood stage 34 days, and above 50 feet 24 clays. In 1912 it was above flood stage 62 
days, and above 50 feet 29 days. 

The 1912 record was exceeded at Xew Madrid, Mo., 0.5 foot; Memphis. 1.2 feet: Helena. 
Ark., 0.8 foot. 

The long continued rise at Helena was due to the water that escaped through the several 
breaks in the Arkansas levee, returning by way of the St. Francis River which joins the main 
stream 8 miles above Helena, Ark. Conditions somewhat similar obtained at Xew Madrid 
where the river rose for five days after it became stationary at Cairo and showed no material 
change for six clays thereafter. 11 days in all. This was caused by the overflow water from 
breaks in levees opposite and above Cairo passing south via St. Johns Bayou and the swamp 
region in southeast Missouri and rejoining the Mississippi above Xew Madrid. 

A matter of special interest in connection with floods in the section of the river between 
Cairo and Helena is the change in gage relation between Cairo and stations north of Helena. 
At the time of the great flood of 1882, the excess in gage height at Cairo, above Memphis, was 
16.8 feet, this being near the average difference between the two gages previous to the con- 
traction of the protecting levees in this district. In 1897 this difference was reduced to 14.5 
feet; in 1903. the difference was 10.5; in 1907, 10; in 1912, it had dropped to 8.7 feet, and in 

1913, to 8.2 feet. While it is possible that had no break occurred in the levee, the river at 
Memphis would have reached a foot higher than it actually did, it is equally possible, had no 
break occurred at Cairo, that gage would have indicated a higher stage, therefore it seems 
probable that under present conditions the difference between Cairo and Memphis will continue 
close to 8 feet. 

At Helena the effect of levee construction in Mississippi was first seen in 1897, when the 
difference between that station and Cairo was changed from —4.6 feet to +0.2 feet, an elevation 
of the flood plane of practically 5 feet. Since that time Helena has ranged from one to five- 
tenths above the Cairo reading. 

Crevasses. — On March 31. at 5 p. m.. the levee at Columbus. Ky.. gave way and caused 
that city to be flooded to a depth of from 5 to 10 feet. As all hope of saving the levee had 
been abandoned early in the day. the people were warned to leave and as a result only a few 
families remained when the flood waters entered the town. 



FLOOD IN THE MISSISSIPPI RIVER, APRIL, 1913, CAIRO, ILL., TO HELENA, ARK. 91 

On April 4, at 12.40 p. m. the levee protecting West Hickman, Ivy., gave way and in an 
hour and a half all of that portion of Hickman was flooded to a depth of from 4 to 15 feet. 
In hundreds of houses the water reached to the top of the windows and every store building had 
water halfway to the ceiling. About 2,000 residents of the city took refuge on the high ground 
near by. 

On April 5, at 8 a. m., a portion of the levee in North Memphis bordering Bayou Gayoso 
gave way and later on other portions of the same levee went out, resulting in severely flooding 
about 20 city blocks and drove 1,000 or more families from their homes and caused many of 
the manufacturing concerns to close. Owing to the alertness on the part of the city officials 
in warning people in the threatened district, there was no loss of life, and much valuable 
property was saved. 

On April 9, at 5.30 p. m., the St. Francis levee near the town of Wilson, Ark. (at a place 
called Golden Lake), broke after two Aveeks of the most persistent high-water fight on record. 
At the time the break occurred the water was splashing over the sandbag topping at intervals 
of 500 feet for about three-fourths of a mile. Following this break, another occurred at 
Random Shot, 4 miles below Golden Lake. This occurred at about 11 p. m., April 9. 

On April 10, at 2 a. m., a break occurred in the Arkansas levee near Graves Bayou, Ark., 
23 miles below Memphis. The two breaks in the levee near Wilson, Ark., and the one at 
Graves Bayou, relieved the strain on the whole St. Francis system and prevented breaks in 
other portions of the levee that at the time were in great danger. The waters escaping through 
these crevasses, as they moved southward through the St. Francis River and the bayous and 
swamps draining into it by which they would eventually, reach the Mississippi River near 
Helena, flooded to some degree parts of seven counties in Arkansas. The northeastern portion 
of Crittenden County was saved from overflow from the Wilson breaks by a low ridge near 
its northern boundary. This entire county was inundated in 1912. The eastern portion of 
Mississippi County, Ark., was also saved from overflow, but with these exceptions, the inunda- 
tion in that portion of the St. Francis Basin lying in Arkansas was as complete as in any 
previous high water in recent years. The counties of Poinsett and Cross suffered severely, 
the water being from 2 to 3 feet higher in some sections than in 1912, while in St. Francis, 
Lee, and the greater portion of Crittenden Counties the flood was about one-half foot below 
1912. The St. Francis River at Madison, St. Francis County, Ark., came to a stand on April 
22 at a stage of 40.9 feet, which is 0.9 foot below the 1912 record at that place. 

The St. Francis levee district, which comprises the St. Francis Basin proper, extends from 
Point Pleasant, Mo., southward to Helena, Ark., being bounded on the east by the Mississippi 
River and on the west by Crawleys Ridge. In this district there are 3,200 squares miles subject 
to overflow, 2.500 square miles being in Arkansas and TOO in Missouri. It is estimated that 75 
per cent of the Arkansas lands in this district subject to overflow were flooded, and those 
located in Missouri comprise about 20 per cent of the 700 square miles, making a total in the 
St. Francis Basin of 2,015 square miles of flooded land. This is about 705 square miles less 
than in 1912. 

The overflowed territory in Missouri from the high ground near New Madrid, Mo., to 
about opposite Cairo, 111., comprises 439 square miles, making a total area on the right bank of 
the Mississippi River, between Cairo and Helena, directly flooded from that stream of 2,454 
square miles, an area 400 square miles greater than the State of Delaware. Fortunately at 
the beginning of the flood period the St, Francis River was at a low stage, which permitted 
much of the overflow water from the upper districts to be carried off before the crevasses lower 
down occurred. Then again the recently constructed drainage canals in the northern portion 
of the basin aided materially in carrying off the waters, especially those that found their way 
into the basin through the crevasses in the vicinity of Cairo. Considering the great volume of 
water that entered the St. Francis Basin the land was cleared in a remarkably short time. 
Before the end of April farming operations were well under way in most of the northern 
counties, and by May 5 practically all portions of the basin were free of water. 



92 THE FLOODS OF 1913. 

Flood warnings. — On March 26 the public was advised by a special warning issued by the 
local office of the Weather Bureau to prepare for a severe flood, and the statement was made 
that a stage exceeding 40 feet was certain at Memphis, and that the stage at Helena would 
exceed 50 feet. On the following day, March 27, the following bulletin was issued and widely 
distributed throughout this district: 

The river at Memphis will reach flood stage (35 feet) hy Saturday or Sunday, and 40 feet in the next 
five or six days. The water now in sight is sufficient to give Memphis considerably more than 40 feet, and it 
is possible the maximum stage may approximate the 1912 record of 45.3 feet. At Helena a stage considerably 
above 50 feet is indicated. 

Immediately following the publication of these warnings the several levee boards, Gov- 
ernment engineers in charge of river improvement under the Mississippi River Commission, 
and the city of Memphis began to prepare for the expected flood. The writer of this report 
can confidently state that during no previous flood period during the past 20 years or more 
have the preparations for a high-water fight been as carefully planned or more successfully 
carried out than during the flood of 1913. Much credit is due Maj. E. M. Markham, in charge 
of the first and second districts, Mississippi River Commission, and his assistants, and Mr. B. G. 
Covington, chief engineer for the St. Francis levee board, in holding the levees intact when 
the river had reached a stage from 1 to 2 feet above the grade they were intended to withstand, 
and in not relaxing their efforts until all danger had passed. 

As soon as this office announced that the flood would probably approximate the one that 
occurred in 1912, the St. Francis levee board immediately dispatched crews of men to Wilson, 
Ark., to strengthen the levee at that place, a crevasse having occurred there in 1912, and a 
weakness had developed during the January rise of this year. Also men and teams were 
distributed at convenient localities along the entire line. Government steamers and quarter 
boats were ordered assembled and supplied with stores, bags, and other material in order to be 
ready for an emergency call. In Memphis every available man in the city service was set at 
work strengthening and enlarging the levee bordering Bayou Gayoso. This levee is for the 
protection of north Memphis from back water in Wolf River, which in turn causes the bayou 
to overflow and flood that portion of the city whenever the Mississippi reaches a stage of 40 
feet or over on the Memphis gage. At this time telephone and telegraph messages were pour- 
ing into this office from all directions and in great numbers. Many of these messages contained 
requests for information in regard to the advisability of moving stock and other property from 
certain localities; others inquiring the date the river would reach a certain stage in order that 
they could arrange for moving before the water reached them. Railroad officials, both in 
Memphis and in distant cities, sent in requests to be daily advised as to conditions in this 
district and future prospects. 

On April 2 an additional warning was issued that the river at Memphis would reach 46 feet, 
and 55 feet or over at Helena. As the North Memphis levee was built to withstand only 42 
feet on the Memphis gage, the city engineer department was strongly advised to prepare 
for 46 feet in the next 10 days. Accordingly hundreds of men and teams were set at work 
in night and day shifts in an effort to raise the levee to the proper grade. The Memphis 
Light & Power Co. and the Memphis Gas & Electric Co. began raising their private levees to 
higher levels. On April 7 the prediction was made that the stage at Memphis would exceed 
46 feet, the amount depending on the stability of the Arkansas levee. On April 11 the city 
of Helena was advised that the maximum stage at that place would be 55.5 feet. The levee 
at that place had been raised to a grade that it was thought would stand 56 feet. On April 22 
the river at Helena came to a stand at a stage of 55.3 feet, 0.2 of a foot below the estimate 
made 11 days previous. 

Owing to the fact that the water along the Arkansas front was somewhat above the stage 
the .levees were built to withstand, and in several long stretches the water was being held back 
by bags of sand, it was not possible to make an exact forecast as to the ultimate height of the 
flood at Memphis. However, the estimate of 46 feet made on April 2 was given 8 days in 
advance of the crest stage of 46.5 feet, and with the estimate of 45 feet 12 days in advance 
gave the people reasonable time in which to prepare for the coming flood. 



REPORT ON FLOODS OF 1913— VICKSBURG RIVER DISTRICT. 



By William E. Barron. Section Director. 

The floods during the spring of 1913 in the Vicksburg River district consisted of two 
distinct rises, the second of which was the more important. 

The Mississippi River began to rise in this district on January 6 with comparatively low 
stages prevailing, in which respect this rise was much like the rises of February-March, 
1909, and of January-February, 1910. Flood stages were passed at Arkansas City, Ark., 
47 feet, on January 30; at Greenville, Miss., 42 feet, on February 2; and at Vicksburg, 45 
feet, on February 1. The maximum stages were 50 feet at Arkansas City and 43.7 feet at 
Greenville on February 11 and 12, and 49 feet at Vicksburg on February 16-18, inclusive. 
The river fell below flood stage at Arkansas City on February 19, at Greenville on February 18, 
and at Vicksburg on February 26. 

The river continued falling until March 4, when a slow rise set in that gave stages of 35.5 
feet at Arkansas City and of 29.2 feet at Greenville on March 12 and 13, and of 38.3 feet at 
Vicksburg on March 14, after which the river again declined to 31.1 feet at Arkansas City, 
March 19; 25.1 feet at Greenville, March 20; and 34 feet at Vicksburg, March 22, by which time 
the Ohio at Cairo, 111., had been rising for a full week prior to the phenomenal rains of March 
23-27 over the Ohio watershed. 

The rise in the Vicksburg district was continuous from this time until the stages began to 
be affected by breaks in the levees. Flood stages were reached at Arkansas City April 5, and 
at Greenville and Vicksburg April 8. The maximum stages were as follows: Arkansas City 
55.1 feet, April 21-25, inclusive ; Greenville, 50.4 feet, April 21 ; Lake Providence, La., 48 feet, 
on the same date: and Vicksburg, 52.3 feet, April 27 and 28. These stages are three-tenths, 
two-tenths, and one-tenth of a foot, respectively, less than the record-breaking stages of 1912 
at Arkansas City, Greenville, and Lake Providence. At Vicksburg the crest exceeded that of 
1912 by two-tenths of a foot, but was two-tenths lower than the high-water mark of 1897. The 
decline was rapid, and the river passed below flood stage at Arkansas City May 9, at Green- 
ville May 8, and at Vicksburg during May 15. 

Frequent rains during January caused a gradual rise in the Yazoo River. At Swan Lake, 
Tallahatchie County, Miss., it passed above flood stage, 24 feet, on January 16, and reached a 
maximum stage of 29.3 feet January 30-February 5. As the country drained by the upper 
Yazoo is flat, and as there were copious rains from time to time, the fall was slow, and by 
February 26 the water had only receded to 26.8 feet, 2.8 above flood stage. For the next four 
weeks the stage fluctuated slowly between 27.2 and 26.5 feet. Beginning with the latter 
reading on March 25, as a result of rains over the watershed on March 20 and 21 and March 
23-27, there was another rise to 29.3 feet on April 7 and 8, after which there was a slow fall, 
the river finally going below flood stage during May 7. 

At Greenwood, Leflore County, Miss., on the Yazoo, the highest stage reached on the first 
rise was 35.6 feet, 2.4 feet below the established flood stage, February 12. The highest stage 
on the second rise was 33.3 feet, April 16 and 17. 

At Yazoo City, Yazoo County, Miss., on the Yazoo, the river passed above flood stage, 25 
feet, during February 6, and reached a maximum stage of 28.9 feet, February 19-23, inclusive. 



94 THE FLOODS OF 1913. 

The river fell below flood stage at this point on March 27. over a month after the Mississippi 
passed below flood stage at Vicksburg. The water had only fallen to a stage of 24.3 feet on 
April 3, when the second rise set in. Flood stage was reached again on April 5, and the maxi- 
mum stage. 29.8 feet, occurred May 2-4, inclusive. From this time the river fell and passed 
below flood stage during May 19. 

The Yazoo was above flood stage at Swan Lake continuously for 112 days, and at Yazoo 
City at or above flood stage a total period of 93 days, although not continuous. 

There were five crevasses in the Vicksburg district. One, the break in the levee near 
Beulah, Bolivar County, Miss., occurred during the first rise, and the others during the second 
rise. 

The most important crevasse of the second rise in the Vicksburg district was the one in the 
Skipworth levee, about 3^ miles north of Mayersville, Issaquena Count}'. Miss. The water from 
the Skipworth crevasse passed out along Steels Bayou and Deer Creek and overspread an area 
of 308 square miles that otherwise would not have suffered in Issaquena, Sharkey, and Wash- 
ington Counties, and increased the depth of overflow in the lower portion of the delta where the 
lands were already covered with backwater that had come in between the lower end of the levee 
system at Eagle Bend, Warren County, and the mouth of the Yazoo at Vicksburg. The total 
area flooded in the lower Yazoo district from backwater and the Beulah and Skipworth 
crevasses was 2,235 square miles, as against 2.723 square miles in 1912. 

The overflow from Beulah crevasse was 702 square miles as compared with 1.498 square 
miles in 1912. 

Three breaks occurred in the levees along the right bank of the Mississippi in the Vicksburg 
district. The effect of these breaks was merely to increase the depth and extent of the back- 
water overflow above the mouths of the White and Arkansas Rivers and Cypress Creek. The 
total area flooded* in the White River district, above the mouth of the White River, was 300 
square miles, out of 910 square miles subject to overflow. 

On the right bank south of the Arkansas River the whole of this district is protected by 
levees, with the exception of a gap of 2.9 miles between the levees of the Arkansas and those of 
the Mississippi, through which Cypress Creek flows into the Mississippi. South of Cypress 
Creek there is a ridge of comparatively high ground, about 12 miles long, known as Amos 
Bayou Ridge. Under present conditions water begins to flow over this ridge into the upper 
Tensas Basin at a stage between 51 and 52 feet on the Arkansas City gage. It is estimated that 
this discharge, both in 1912 and 1913, reached as much as 200,000 cubic feet per second. The 
water discharged over this ridge spread southward between the Mississippi River levees and 
the McGehee and Alexandria line of the St. Louis, Iron Mountain & Southern Railway, follow- 
ing the courses of the Boeuf River and Bayou La Fourche into the Ouachita, and of the Bayou 
Macon into the Tensas, and thence into the Black and lower Red Rivers. At Arkansas City, 
Ark., the water from this overflow reached a height equivalent to a stage of 47 feet on the river 
gage at that point. The total area overflowed along the right bank in this district south of the 
mouth of the Arkansas River was 1,591 square miles, as compared with 2,270 square miles 
submerged in 1912. 

The public in this district was kept advised of the stages of water to be expected at the 
several stations, even prior to the issuing of warnings of flood stages. Flood warnings were 
issued for Arkansas City and Vicksburg on January 18, and for Greenville on January 23, 10 
to 14 daj's in advance, and further advice was furnished during the progress of the rise. The 
occurrence of the crevasse at Lake Beulah on January 25, before flood stages were reached, 
presented a problematical condition and at first it was difficult to tell just what the effect would 
be. The crevasse reduced the flood height but slightly at Arkansas City and Greenville, and 
prolonged the rise at Vicksburg about three days on account of the return of the crevasse water 
through the lower Yazoo, and a forecast was made on January 30 covering the situation. 

On March 27, as soon as it was seen that the heavy rains of March 23-27 over the Ohio 
watershed would produce a marked rise in the Mississippi, warnings were issued that flood 
stages would be reached in this district April 5 to 8. This forecast was verified to the day. 



REPORT ON FLOODS OF 1913 VICKSBURG RIVER DISTRICT. 95 

During the early part of the rise the stages forecast were modified as conditions warranted, 
but it was not believed that the stages of 1912 were probable in this district on account of the 
lower initial stages. By April 1 the prediction was made that a stage of 54 to 55 feet would 
be reached at Arkansas City, 48 to 49 feet at Greenville, and 51 to 52 feet at Vicksburg if the 
levees held. On April 7, owing to the prolonged high stage of 54.7 feet at Cairo, 111., the fore- 
cast was changed to 55.5 to 56 feet at Arkansas City April 17 to 18, 49 to 50 feet at Greenville 
April 18 to 19, and 52 to 53 feet at Vicksburg April 23 to 24. On April 10 the public was 
informed that the recent heavy rains over Arkansas would intensify flood conditions, but that 
the crest would be delayed b} ? crevasses that had occurred in the St. Francis levees. On April 
13 the crevasse water from the St. Francis Basin began to return to the main stream at 
Helena. Ark. The effect that this amount of water would have on the stages below Helena was 
considered and the following special warning was issued : 

The Mississippi will pass 55 feet at Arkansas City, 50 feet at Greenville, and 50.5 feet at Vicksburg by 
April 17 or IS, after which a slow rise will continue till late in April or early in May, due to the return of the 
crevasse water from the St. Francis Basin, reaching 56.5 to 57 feet at Arkansas City, 51.5 to 52 feet at 
Greenville, and 53 to 54 feet at Vicksburg if levees hold. 

On April 21, on account of breaks that had occurred in the White River district, which 
caused an increase in the discharge of water over Amos Bayou Ridge, the estimate was changed 
to 55.5 to 56 feet at Arkansas City, 51 to 51.5 feet at Greenville, and 52.5 to 53 feet at Vicks- 
burg, providing that there were no further breaks in the levees. 

The crevasse in the Skipworth levee occurred that afternoon, and as soon as definite infor- 
mation was received at this office a special forecast was issued that the effect of this crevasse 
would be to relieve Greenville and Arkansas City and to bring temporary relief to Vicksburg. 
By the next morning the gage at Lake Providence, La., 12 miles below the crevasse, showed a 
fall of 1.2 feet, while at Greenville, 52 miles above it, there was a fall of two-tenths of a foot, 
and at Arkansas City and Vicksburg the rise had been checked, although the river was still 
rising at Helena, Ark. The fall was continuous after the 21st at Greenville and after the 25th 
at Arkansas City. By the latter date, as a result of heavy rains in the Aucinity and because 
of the return of the crevasse water, the river was again rising at Vicksburg. and a forecast 
was issued that a stage of 52.5 feet would be reached. As a restdt, however, of the crevasse 
that had occurred in the levee at Lake St. John. La., on the night of April 26, the river was 
stationary at Vicksburg at a stage of 52.3 feet from the morning of the 27th to the evening of 
the 28th, and then began to fall slowly. 

After the river began to fall the vital question was when the water would pass beloAv 
flood stage. On May 5 the public was informed that the Mississippi River would pass below 
flood stage at Arkansas City and Greenville on the 8th or 9th. and at Vicksburg by Maj^ 18, 
barring heavy rains. On May 7 further advice was issued that the Mississippi River would 
pass below flood stage at Vicksburg by Ma} 7 15 and that the Yazoo River at Yazoo City would 
pass below flood stage by May 22 or 23, barring heavy rains. All of these predictions were 
verified. In fact, it Avas possible on the 17th to advance the date at Yazoo City to May 20. 

During both rises ample warnings were issued for points on the Yazoo River as occasion 
required. 

The warnings and other information relative to the flood issued at this office were given 
wide distribution. During the critical period the River Bulletin reached a circulation of 1,100 
copies. The river conditions were also transmitted by telegraph and telephone without expense 
to the Government to the levee boards at Greenville, Miss., Tallulah, Lake Providence, and 
St. Joseph, La., to the Cotton & Merchants Exchange, at Natchez, Miss., and to other interested 
parties over the district. The board of commissioners of the fifth Louisiana levee district at 
Tallulah duplicated the river bulletin thus given to them and circulated it to 20 addresses along 
the line of the St. Louis. Iron Mountain & Southern Railway. 

The losses to residents of the districts submerged during the first rise were not great, 
because the overflow occurred after last year's crops had been nearly gathered and before 
planting had begun, because the waters came up gradually, as the only levee break, that at 



96 THE FLOODS OF 1913. 

Beulah, occurred before the river had reached a height sufficient to cause a current of great 
force, and because the people knew they were unprotected and were expecting this overflow. 

The largest losses fell on the railroad companies. The Riverside division of the Yazoo 
& Mississippi Valley Railroad Co. was out of commission between Benoit and Beulah, Miss., 
from January 26 to February 26, practically the entire roadbed being washed from under the 
track between those places. The main line of the same railroad sustained a bad washout 
between Valley Park and Smedes, Miss., and was closed to all traffic from February 19 to 22, 
inclusive, while service on the branch line from Kelso to Silver City was discontinued on Jan- 
uary 31. The main line of the Southern Railway was under water between Holly Ridge 
and Elizabeth, Miss., from January 29 to March 5. The Napanee branch was also put out 
of commission for some time, as was also the Richey branch from Delta City to Richey, Miss. 

The loss from the second rise was much greater, owing- to its later occurrence, to the 
higher stages, and to the consequent wider extent of the overflow. When the crevasse occurred 
in the Skipwith levee, couriers were sent in all directions to warn the inhabitants of the terri- 
tory in advance of the flood waters. The United States Army relief forces that were stationed 
in readiness at Vicksburg dispatched boats, men, and provisions to the scene, and heroic efforts 
were put forth to protect human life and care for live stock. It is estimated that there were 
10,000 people rendered homeless by this disaster, but it is believed that there were no lives lost, 
although at first it was reported that there were two. There was much exposure to refugees 
on account of heavy rains on April 23-24. Buildings on the immediate front of the wave were 
carried away, as were practically all the bridges over Steels Bayou in Issaquena County. The 
more important towns and villages affected were Mayersville, Rolling Fork, Grace, Blanton, 
and Carey. Regular train service on the main line and on the Riverside division of the 
Yazoo & Mississippi Valley Railroad was discontinued on April 22. The entire line between 
Vicksburg and Kelso and the east side of the embankment between Kelso and Rolling Fork 
was already covered with backwater, which acted as a buffer, and the railroad company was 
able to hold most of their track in place when the crevasse water came, so that they were able 
to effect an early resumption of service on May 12. Service on the Silver City branch of this 
railroad, discontinued January 31, was not resumed until June 2. 

While some of the lands were damaged by erosion, particularly in the immediate vicinity 
of the crevasse, where a deep hole covering 102 acres was washed out, many planters believe 
the overflow a benefit on account of the soil-enriching sediment, which in many places was 
deposited 2 to 5 inches deep. As the water receded, the planting of cotton was begun on May 
15 with favorable prospects. 

The most important towns affected by the backwater in Arkansas and Louisiana, were 
Arkansas City and Lake Village, both in Arkansas. In this territory there were about 10,000 
people who were temporarily forced from their homes. 

Reports of money losses have been furnished from portions of the flooded districts only, 
and these vary with the viewpoint and temperament of the persons reporting. Conservative 
figures based on all these reports, giving the losses exclusive of those to railroads, telegraph 
and telephone companies, are submitted. The figures given include losses in Vicksburg along 
the river front, where no permanent levees are maintained, and where the losses to buildings 
and contents are placed at $15,000, loss by suspension of business, $11,000, and where the warn- 
ings were undoubtedly instrumental in saving $200,000 worth of property. Probably $15,000 
to $20,000 were expended for local protection. 

The cost of high-water protection over the district will amount to hundreds of thousands 
of dollars, to say nothing of the cost of building and maintaining the levees. Had it not been 
for the long vigil and the tireless efforts kept up throughout the critical period, the losses 
would have been multiplied many times. A grave disaster was narrowly averted at Greenville, 
Miss., as late as April 30. 

The loss to prospective crops may hardly be computed with any certainty. If the season 
is otherwise favorable, the loss may be inconsiderable, while, if unfavorable, it may be even 
greater than the amount estimated! 



COMPARATIVE STAGES IN THE ARKANSAS AND WHITE RIVERS IN 
CONNECTION WITH THE SPRING FLOODS OF 1912 AND 1913 IN THE 
MISSISSIPPI RIVER. 



By H. F. Alciatore. Section Director, Little Rock, A rk. 



In 1912 the spring rise in the Mississippi River at Arkansas City, Ark., about 37 miles 
below the mouth of the Arkansas River, began about March 21. The flood stage was reached 
eight days later, and the crest stage, 55.4 feet, on April 12. 

The volume of water _that passed from the Arkansas River into the Mississippi River is 
indicated by the stages in the Arkansas River at Pine Bluff, Ark., during March and the first 
week in April. Pine Bluff is 109 miles from the mouth of the river. The Arkansas began to 
rise on March 23 ; the stage was then 14.6 feet. There was a continuous rise until April 4. The 
flood stage was reached on April 3 (25.7 feet) , and the crest of the flood occurred on the 4th, the 
gage reading being 26.2 feet, which was 1.5 feet above the flood level. After the 4th the water 
receded steadity until the 26th. when the gage read 12.5 feet. As the crest of the flood in the 
Mississippi did not pass Arkansas City until April 12 (8 days after the Arkansas River flood 
had passed Pine Bluff) it is evident that the waters from the Arkansas had some bearing on 
the flood in the Mississippi below Helena. Ark. 

As to the amount of water contributed by the White River in 1912 in Arkansas the stages 
recorded at Clarendon, Ark. (75 miles from the mouth), may be taken as an index. Relatively 
high stages (27.9 to 28.3 feet) prevailed during the latter half of March. On April 1, the river 
began to rise. The stage was then 28.6 feet. The flood stage was passed on April 6 (stage, 
30.1 feet), and the crest of the flood passed Clarendon on April 14. The stage was 32.6 feet, or 
2.6 feet above the flood plane. After that date the river fell steadily at the rate of one or two 
tenths foot each day. The combined waters of the White and Arkansas Rivers must have had 
appreciable effect on the flood situation in the Mississippi below Helena, Ark., during the early 
part of April, 1912. 

In 1913. the Arkansas River was not so high during March and April as it was during the 
same months of 1912. From the 1st to the 26th of March, 1913, the stages at Pine Bluff, Ark., 
were relatively low, ranging from 8.9 feet on the 21st to 14.4 feet on the 3d. An 8-foot rise 
occurred between the 26th and 31st, the extreme stages recorded being 11.9 feet and 20 feet, 
respectively. 

The torrential rains that occurred in the lower part of the drainage basin of the Arkansas 
(from Little Rock, southward) on April 8 and 9 did not cause floods in any part of the river. 
At Pine Bluff there was a rise of 7.5 feet between the 9th and 14th, the highest stage reported 
being 20.4 feet on the 14th. This was 4.6 feet below the flood stage. It will be seen that in 1913 
the volume of water that passed out of the Arkansas into the Mississippi was relatively small. 

As to the White River, the Clarendon. Ark., records show that there was a steady fall 
during the first 21 days of March, the river reaching a minimum stage of 19.5 feet on the 21st. 
14284°— 13 7 97 



98 THE FLOODS OF 1913. 

On the 22d, the White began to rise. The stage was then 19.8 feet. The flood stage (30 feet) 
was reached on April 13. and the flood crested on the following day (14th), with a stage of 
30.4 feet, which was 0.4 foot above the flood plane. The river remained stationary during the 
loth and began to fair on the 16th. It fell slowly, however, the total fall during the last 15 
days of the month having been only 1.7 feet. 

The spring rise in the Mississippi at Arkansas City in 1913 began about March 20. but 
the flood stage was not reached until April G, 17 days later. The crest of this flood occurred 
April 22, with a maximum stage of 55.1 feet. 

It will be noted that prior to the occurrence of the spring flood of 1912 in the Mississippi 
both the Arkansas and White Rivers had reached higher stages than during the corresponding 
period preceding the 1913 flood, the high-water stages for those years being 26.2 and 20.4 feet, 
respectively, at Pine Bluff, on the Arkansas, and 32.6 and 30. 4 feet, respectively, at Clarendon, 
on the White River; yet the high-water stages recorded at Arkansas City, on the Mississippi, 
were practically the same — i. e.. 55.4 and 55.1 feet, respectively. 

NOTES ON THE SPRING FLOOD OF 1913 ON THE ARKANSAS SIDE OF THE MISSISSIPPI RTVER. 

In extent and severity the flood of April, 1913, on the Arkansas side of the Mississippi 
River differed but little from that of 1912. This year the warnings of the Weather Bureau 
regarding the flood were heeded without question, and as they proved both timely and accurate 
there was no loss of life. 

The first effects of the flood were felt on April 9. The flood w-aters passing through breaks 
in the levees near Graves Bayou and Wilson, Ark., covered portions of Mississippi County and 
spread southward over parts of Poinsett, Crittenden, St. Francis, and Lee Counties. Press 
dispatches stated that on April 9 several refugees had sought shelter on higher ground at 
Forrest City, St. Francis County, and in neighboring towns. Railroads west of Bridge 
Junction. Ark. (opposite Memphis, Tenn.), were practically out of business. Mails from the 
East were being transferred across the Mississippi River at Helena, Ark. By the 11th the 
flood had spread over the lower St. Francis Basin, and was making its way toward the 
Mississippi through the St. Francis River. About 2,000 refugees were camping on high ground 
near Wilson. Mississippi County. On the 12th, Marked Tree, Poinsett County, was flooded to 
a depth of 2 to 5 feet. The flood had reached Deckerville and Turrell, Ark. About 1.000 
flood sufferers were being cared for at Marianna, Ark. On the 13th the towns of Shawnee 
Village, Joiner, Bassett, Lepanto, Tyronza, and Deckerville, Ark., were submerged, and the 
water was higher than in 1912. The St. Francis River at Marked Tree had risen 19 inches 
in the last 24 hours. About 200 negroes were rescued at Watson. About 2,000 refugees from 
the eastern portions of Crittenden County were camped at Wynne. Edmonson, Manila, and 
Clarksdale were under water. A telegram was sent from Osceola, Ark., on April 14 to the 
governor of Arkansas that the flood waters from the breaks in the levee at Wilson were about 
10 feet in depth in the southern. part of Mississippi County, and almost reached Osceola; great 
suffering and destruction were reported in that section. On the 15th the Laconia Circle levee 
in Desha County, near the mouth of the White River, broke and soon afterwards about 18.000 
acres of farming lands were under water. 

The flood near Wilson, Mississippi County, was said to be about 2 feet higher than ever 
known. Marked Tree was still under water. On the 16th backwater from the Laconia Circle 
break covered the streets of Arkansas City to a depth of several inches, but the levee at that 
place was intact. Railroad traffic to the south had been suspended, the tracks near Valley 
and Lake Village being under water. The northeastern part of Chicot County was submerged 
by the waters flowing through a gap in the levee on Amos Bayou. The road between Luna 
and Lake Village was flooded. By the 19th the St. Francis section was under water from 
Hulbert. Crittenden County, westward to Madison, St. Francis County, a distance of about 30 
miles. The Rock Island trains were running through water ranging; from 3 to 20 inches in 



COMPARATIVE STAGES IN THE ARKANSAS AND WHITE RIVERS. 99 

depth. The water from the L'Anguille River was running over the Rock Island Railroad trestle 
near Palestine. Ark., and was overflowing the "dump" to a depth of 12 to 14 inches. On 
April 20 the railroad tracks near Forrest City were submerged. At the L'Anguille River 
crossing trains were running through water for a distance of about 1 mile, the car steps 
being hidden under the water. At TIeth, on the 21st, the railroad tracks were 2 feet under 
water, and the station platform had floated away. The Rock Island tracks from Round Pond 
to Proctor were under water, and the Iron Mountain tracks between Felton and Canaan were 
submerged to a depth of more than 2 feet in some places. The city authorities at Lake Village 
had issued a call for rations for 1.000 refugees. The St. Francis levee had broken at a point 
about 3 miies below Whitehall, and the water inside of the levee was only about 3 inches 
below the water on the river side. Many telegraph and telephone poles along the Rock Island 
road had been damaged or swept away. 

On April 22 the writer left Memphis, Tenn.. about 7 a. m. on a Rock Island train bound 
for Little Rock, Ark. From Hulbert, Ark., a few miles west of the Mississippi River, to Pales- 
tine, Ark., a distance of about 50 miles, he observed that the tracks were under water to a 
depth ranging from a few inches to more than 2 feet over the greater part of the distance. A 
number of telegraph and telephone poles had been washed away and were floating upon the 
water on each side of the railroad track. In some places the flood water had almost reached 
the lower cross-arms on poles that were still standing. Xearlleth a high-water mark which had 
been painted upon a telegraph pole in 1912 w T as only 5 or G inches above the flood waters of 
1913 on April 22. Press reports of that date indicated that the bottom lands between the 
Mississippi River and the LAnguille River, 1 a distance of 25 to -10 miles, were under water. 
At Marianna. Ark., on the St. Francis, the river gage at 7 p. m. showed a stage of 47.4 feet. 
which was about one-half foot below the high-water mark of 1912. 

On April 23, 00,000 rations were distributed by the Federal Government among 3,000 
refugees at Forrest City. The St. Francis River at Madison had fallen about 2 inches, the 
stage being 40.7 feet. The LAnguille River at Palestine had come to a stand. The flood 
situation was beginning to show improvement everywhere. On the 24th the backwater at 
Arkansas City had risen about 3 inches, but half of this rise was due to excessive rains. The 
water was 6 to 8 inches higher than it was in 1912 during the spring flood. At Madison, on 
St. Francis, the flood had receded about 2 inches. All trains over the Rock Island, except Xos. 
45 and 4G, were in operation, but all trains were late. By the 28th the water in the streets of 
Arkansas City had fallen about 4 inches. Near McGehee, Desha County, the flood waters had 
receded about 3 inches. On that date the territory under water in eastern Arkansas was to 
13 miles wide and 30 to 40 miles long. 

At the close of April the rivers were falling at all points in the flooded districts, and the 
situation was improving rapidly. Railroad traffic on lines in eastern Arkansas were running 
practically on schedule time. Large areas were still submerged, but the flood waters were 
receding rapidly, and many flood refugees were returning to their homes. 

The most expensive features of this flood were the strengthening of the levees, the cleaning 
and repairing of stores, dwellings, and farm buildings, and the losses incident to the suspension 
of business throughout the flooded areas. It is impracticable to make a correct estimate of 
the losses at this time, but it is probable that these will exceed $2,000,000. exclusive of the 
damages to the railroads. 

*A tributary of the St. Francis River flowing southward through Poinsett, Cross, and Si. Francis Counties. 



FLOODS IN THE MISSISSIPPI RIVER BELOW VICKSBURG AND IN THE 
ATCHAFALAYA RIVER IN APRIL AND MAY, 1913. 



By I. M. Clink, District Forecaster. 



Stages in the Mississippi River, below Vicksburg and in the Atchafalaya River, on March 
31, 1913. were as follows: Natchez, Miss., 39.2; flood stage, 40 feet. Baton Rouge, 29.2; flood 
stage, 35 feet. Donaldsonville, 22.7; flood stage, 28 feet. New Orleans, 14.5; flood stage. 18 
feet. Melville, 35.3 ; flood stage, 37 feet. Morgan City, 3.7 ; flood stage, 8 feet. On account of 
the high water in the Ohio River and its tributaries flood warnings as follows were issued on 
March 31 : 

If levees bold, the water now in sight indicates 50.5 to 51.5 feet at Natchez, 40 to 41 feet at Baton Rouge, 
32.3 to 33.3 feet at Donaldsonville. and 10.(5 to 20.(5 feet at New Orleans, between April 20 and 30, 1913. 

On account of the changing conditions, such as the fall of rain and the breaking of the 
levees, it was necessary to modify and alter the warnings from time to time during the con- 
tinuance of the flood. Warnings were issued on April 4, 9, 10, 21, 23, and 25, and on May 9, 
12, and 14. the last warning issued, viz, that of May 14, follows: 

The Mississippi River below Vicksburg will fall, but below the mouth of the Red River and in the 
Atchafalaya the changes will be slight for a few days. The water will fall below the flood stage at Natchez 
by May 20. and water will probably cease flowing through the Take St. John crevasse by May 25. 

If no breaks had occurred in the levees below Vicksburg, the maximum stage forecast for 
this district would have been reached. At Natchez a stage of 52.4 feet was reached on April 26 
and 27. which is the highest stage ever recorded at that place. The highest stages at other 
points were as follows: Baton Rouge, 41.3 feet, on May 9; Donaldsonville, 32.7 feet, on May 8 
and 9 ; New Orleans, 20.5 feet, on May 8 ; Simmesport, 40.9 feet, on May 9 : and Melville, 41.7 
feet, on April 24. 

The water was rising rapidly at Natchez when the levee broke 26 miles above that place on 
April 27. and but for that break there would have been a further rise of at least 2 feet which 
would have given the stage forecast. If southerty instead of northerly winds had prevailed 
stages below the mouth of Red River would have been 1 to 2 feet higher than occurred, because 
southerly winds would have retarded the discharge and delayed the flood crest until the 
crevasse water would have returned and united with the flood crest in the main stream. 
However, the prevailing northerly winds caused the flood crest in the river proper to pass out 
into the Gulf of Mexico before the crevasse, water returned through the Red River. 

The following table shows the period during which the water was at or above the flood 
stage: 



Natchez 

Baton Kouj;o-- 
Donaldsonville 
New Orleans . . 
Simmesport. . . 
Melville 



Period above flood 
stage. 



Apr. 11 to May 18. 
Apr. 14 to May 25 . 
Apr. 15 to May 22. 
Apr. 17 to May 24. 
Apr. 17 to May 25. 
Apr. 9 to May 26.. 



Totalnum- 
ber of days 
above flood 



37 
41 
37 
37 
38 
47 



101 



102 THE FLOODS OF 1913. 

Action taken as a result of the warnings. — Warnings were distributed to all post offices in 
the lower Mississippi Valley regularly, by mail, and when changes were made they were 
telegraphed to localities affected. When the first warnings were issued on March 31. five weeks 
before the passage of the flood crest, representatives of all interests likely to be affected' by 
high water took prompt and vigorous measures to combat the coming flood. Levees were 
raised and weak places strengthened ; barges and railroad cars were loaded with materials for 
making repairs and placed in convenient locations ready to be rushed on short notice to any 
weak spot which might develop. These precautionary measures proved to be of great value and 
many threatened breaks in the main levees were stopped by prompt and decisive action, thus 
preventing more extensive destruction than occurred. Planters raised and strengthened the 
protection levees around their plantations in anticipation of a possible break in the main levees. 
This action on the part of the planters enabled several of them to save their plantations from 
being flooded. A large percentage of stock was driven out of the bottoms to high ground 
when the warnings were first issued and action was taken to protect household goods and farm 
products which had been housed. 

A special to the Daily Picayune, New Orleans, La., dated Jackson, Miss.. April 11. 1913, 
in speaking of the flood says : 

Because his warnings in previous years have been so accurate, the people residing in the low lands are 
giving heed promptly to the special flood forecast issued by District Observer Cline, of the Weather Bureau, at 
New Orleans. La., and are preparing to get live stock to the hills and otherwise safeguard themselves against 
loss. 

When the warning for 54.5 feet was issued for Natchez on April 25, planters, believing 
that the levees could not be held against such a. flood, moved out the remainder of their live stock 
so that when the crevasse occurred on April 27 nearW all live stock were out of danger and other 
precautions had been taken, which resulted in the saving of property of great value which 
otherwise would have been destroyed by the crevasse water. 

Crevasses in the New Orleans district. — Only two crevasses in the main levees occurred 
in this district. 

April 25 a crevasse occurred on the left bank of the Atchafalaya River about 8 miles below 
Melville. No attempt was made to close the crevasse. The water from this crevasse returned to 
the Atchafalaya River below the end of the levee line. The water from the crevasse overflowed 
parts of Pointe Coupee and Iberville Parishes. 

April 27 a crevasse occurred on the right bank of the Mississippi River at Lake St. John, 
about 26 miles above Natchez and near the south end of Tensas Parish. The water from this 
crevasse overflowed Concordia and parts of Tensas, Catahoula. Rapides, and Avoyelles Parishes. 
Water ceased flowing through the crevasse May 23, and the crevasse water disappeared by 
June 20. 

A break occurred in a private levee opposite Briers, La., on the left bank of the Mississippi 
River on April 1C, flooding about 3,500 acres of farming land with back water. 

April 24. a break was barely averted at Remy, La., about 40 miles above New Orleans. The 
old levee caved rapidly, and for a time it appeared that the new levee could not be gotten into 
shape to prevent a bad crevasse. By having materials at hand, and by hard work da}^ and night, 
a crevasse was prevented. (See fig. 13.) 

May 1 a cave occurred in the main levee at Poydras Plantation, on the left bank of the 
Mi-sissippi River, about 10 miles below New Orleans. Prompt action in this case also prevented 
a serious crevasse. 

Area -flooded and damage resulting therefrom. — About 1,026,000 acres of land was flooded. 
of which 255.400 acres is agriculture land. Growing crops were destroyed, and the overflow 
made it necessaiy to replant all crops. Cotton, cane, and gardens have been, or will be. 
replanted. Houses, farm buildings, household effects, "public roads and bridges, and railroads 
suffered damages in varying degrees. No lives were lost as a direct result of the flood, but 
several lives were lost in accidents indirectly due to the flood. The steamer Concordia, while 





■i 






L 


1 




a. 






w 


L 


i ^mmm 




• 




1 




n! 


ft 


i 


Itiv : \ 




UM 






■ jV\'' 31m 






Raft.- 


i 


i 






*! 






| 








tS . ; # ^Hj 






tflBBHMf ^i^B P^m^^^H ^^^ 




flillk^M 




-H 




! 
m 


«S 




^■■■fli BHKStLui'i^BWRrfBi ^^ifc T * 


%i 


<l filfi N^fl PrM 


ivy*.,, .'^Jm^j%i 


,4lH^|MMra 


^^iQf*^m 


1 ' -^ *E^rtEiimmr 


♦JjJ^BjM 



FLOODS IN THE MISSISSIPPI RIVER UELOW VICKSBURC, 1913. 103 

engaged in rescue work on May 1, struck the drawbridge of the New Orleans & Northwestern 
Railroad across the Tensas River at Clayton, La., and sank in the river, drowning 22 persons, 
of which number 20 were negroes, mostly women and children. 

From the best source of information available up to the present time the damage resulting 
from the flood is approximately as follows: 

] rem No. 1 : 

(a) Damage to buildings and residences $7,500 

(&) Damage to public roads and bridges G, 250 

Item No. 2 includes farm property, and is to be staled under — 

(a) Loss of crops which may or may not have been housed 240,000 

(6) Loss of prospective crops (255,400 acres 1 ) 225,000 

(c) Loss of live stock and other movable property 12,000 

Item No. 3. Loss due to suspension of business, including wages of employees 75,000 

Item No. 4. Money value of property saved by warnings (actually reported) 015,000 

The money value of property saved by the warnings actually reported does not represent 
more than half the savings as a result of the flood warnings because many correspondents do not 
give amounts. 

There have been no extensions of the levee systems since the flood of 1912. 

1 Not all in cultivation ; planting had been delayed as a result of flood warnings. 



FLOOD IN THE HUDSON RIVER, MARCH 27-28, 1913. 



By George T. Todd, Local Forecaster. 



Flood warnings were issued ar 9 a. m. March 26, 1913,. for Albany, Troy, and vicinity, and 
a forecast was made that the water would reach the flood stage (12 feet) that day and probably 
reach a height of 15 to 1C feet within the next 24 to 36 hours at Albany and a height of 19 to 
20 feet at Troy. A cautionary warning was also given the American Locomotive Works and 
the General Electric Co., at Schenectady. 

On March 27, the heavy rains having continued in the upper tributaries, a second flood 
warning was issued for Albany and Troy, and a forecast was made that the water would 
reach a height of 20 to 21 feet at Albany and 24 to 25 feet at Troy within the next 24 to 30 
hours, and a special forecast was made for the press that the water would reach 19 feet at 
Albany at about 6 p. m. that night. 

The Water at Albany reached a height of 16.8 feet at 8 a. m. on the 27th ; 19 feet at <'» 
p. m., and at 1 p. m. March 28th a maximum stage of 22.4 feet was reached. 

Except for the flood of February 9, 1857, when the water reached a stage of 22.49 feet, 
according to the present datum, and there was a strong ice gorge in the Hudson River just 
below this city, this is the highest stage of which there is any record. 

Before noon of the 27th the rising water had put out the fire in the heaters in the United 
States post-office building in which the Weather Bureau is quartered and the electric current 
for lights and power was shut off about the same time. The office force were compelled to work 
in the rooms without any heat till Monday morning, March 30. 

So far as known, there was only one human life lost during this flood. 

The greatest loss from the flood was occasioned by persons Avho, though doing business near 
the river, were unable to believe that their property would be flooded. When warned they 
would say that the firm had been in business in that section or store for 50, 75, or 100 years 
and that their building had never been flooded before except when there was an ice gorge; 
that they thought this office must be mistaken, and that they did not believe it was possible for 
the water to reach a height of 21 feet or more without the help of an ice gorge. There was 
also a considerable loss through the scarcitj^ of help and the comparatively short time allowed 
for the moving of such a large amount of goods. 

FLOODS IN NEW YORK. 

Mr. Robert E. Horton, consulting hydraulic engineer, Albany, N. Y., in the Engineering 
Record of April 5, 1913, describes the floods in the State elsewhere than in the Hudson water- 
shed. He says: 

Heavy rains which were general throughout the State on March 25, 26, and 27 caused floods of unusual 
magnitude on nearly all streams in New York. Streams in eastern New York reached their highest stages 
early on the morning of March 28. Wherever reliable reports can be obtained the volume of discharge at 

105 



10() THE FLOODS OF 1913. 

the Mood crest was very much greater than for any preceding floods for which there are authentic records. 
The Hood was due to rainfall alone, coming at a time when the rivers were already swollen, the ground 
saturated, and lakes and marshes full. The height of water was not accentuated either by ice gorges or by 
the failure Of dams or reservoirs, yet the stages reached in most places were higher than any hitherto known. 

The average rainfall throughout eastern New York from the morning of March 24 to the morning of 
March 28 was about 1 Inches. The amount varied in different localities from 2 and 2.2 inches at Albany to 
5 inches or more In the southern Adirondack slope. The discharge of the Hudson River increased in 48 hours, 
between the morning of .March 26 and the morning of March 28, from 40,000 to 120.000 second-feet. This was 
at Mechanicsvllle, 18 miles above Albany, where the drainage area is 4,500 square miles. The maximum dis- 
charge of the Mohawk River, which is the chief tributary to the Hudson, is not yet known, but may be safely 
estimated al from 75,000 to 100,000 second-feet, so that the combined flow of the Hudson and Mohawk Rivers 
past Albany was approximately equal in volume to the Niagara River. 

The greatest flood hitherto recorded on the Hudson River, in 1857, reached a volume of 70,000 second-feet 
at Mechanicsvllle, and a stage of 21. 10 feet above mean tide at Albany. The maximum stage at Albany in 
the present flood was 22.4 feet above mean tide at 1 p. m. March 28. Large areas were overflowed in the 
most populous and built-up districts at Albany. Troy, Watervliel, and Green Island on the Hudson River and 
at Schenectady, Fonda. Amsterdam, St. .Tohnsville, and other towns along the Mohawk River. The entire 
Mohawk River Hats from Little Falls to Schenectady, a distance of GO miles, were submerged to the greatest 
depth ever known. 

The present Erie Canal, the four-track system of the New York Central Railroad, trunk telephone and 
telegraph lines, an important State highway, the West Shore Railroad, and for a large portion of the distance 
a double-track trolley system traverse this valley. In addition there are eight locks and movable dams 
recently completed for the canalization of the Mohawk River. Traffic of all kinds was interrupted and con- 
siderable damage was done locally to railroad roadbeds and to the present Erie Canal. All of the new 
Barge Canal locks in this district were completely submerged, but so far as can be learned the Barge Canal 
structures here and elsewhere in the State suffered but little damage. The gates of one of the movable dams 
at Fonda were bent and injured. This was probably due to drift coming down against the gates, making it 
impossible to raise them. 

In the cities and towns great inconvenience was caused to people located within the flooded districts. 
Timely warnings, however, were issued by the local forecaster, Mr. George T. Todd, of the Albany station. 
Many persons heeded these warnings and avoided injury to their property as much as possible. Others refused 
to believe that a flood greater than that of 1857 was coining upon them. They did not clear their cellars or 
move their perishable goods, and suffered serious consequences. In the Hudson River there are many water- 
power dams, some of them constructed of timber or cribwork and a number very old. All of these withstood 
the flood in safety. 

Many bridges were destroyed, including a highway bridge at Glens Falls across the Hudson River, the 
Erie Canal bridge across the Hudson River at Northumberland, the highway bridge across the Hudson River 
at Amsterdam, the highway bridge across the Mohawk River at Herkimer, and the highway bridge across 
West Canada Creek at Trenton Falls. 

Damage to the banks of the Erie Canal in different parts of the State is reported by the department of 
public works as follows: About 100 feet of canal bank washed out at a point opposite the cemetery road 
between Albany and Watervliel : a considerable length of canal bank washed out west of Schenectady, near 
Patersonville ; about 50 feet of canal bank washed out between Locks 35 and 3G, just east of Little Falls; 
about 150 feet of canal bank washed out on the Cayuga and Seneca Canal at Seneca Falls. The last-mentioned 
washout caused the shutting down of various mills in Seneca Falls. 

A number of Erie Canal bridges were injured or destroyed and many of the smaller canal aqueducts were 
damaged. The flood on the Mohawk River reached a stage :> feet above the bottom of the trunk of the main 
Erie Canal Aqueduct crossing the Mohawk River at Crescent. This is a very old stone aqueduct, but in spite 
of the high stage and large volume of water it withstood the flood in safety 

Contractors on the Barge Canal in the Hudson and Mohawk Rivers had their concrete and structural 
work mostly completed and their plants removed from the river sflme time ago. There were many dredges in 
these rivers, but with one exception all kept their anchorages. One dredge, which broke loose from its moor- 
ings drifted downstream about 1 mile and stranded upon an island. Contractors lost considerable property in 
t he way of pipe and other equipment, but the loss is very much less than would have been the case had the flood 
occurred a year or two sooner. 

In view of the Large industrial developments in the valleys of the Mohawk and Hudson Rivers, and in 
view of the sudden rise and unprecedented height reached by this flood, it is extremely remarkable that the 
amount of damage done was not very much greater. The fact that there were so few great individual calami- 
ties is also notable. The damage, great as it is, was very widely distributed, and is made up of an enormous 
number of relatively small items. It is impossible to estimate its amount with any degree of certainty. 

The Oswego River, according to the best reports obtainable, did not suffer so severe a flood as was 
experienced elsewhere in the State. This was undoubtedly due to the great natural storage furnished by the 



FLOODS IN THE HUDSON RIVER, MARCH 27-28, 1913. 107 

finger lakes in central Now York, which are tributary to the Oswego River through the Oneida and Seneca 
Rivers. 

The Genesee River rose to a greater height than has hitherto been known, flooding a large district of the 
city of Rochester and causing extensive damage. The greatest floods of the Genesee River in past years 
were in 1865 and in 1894. The discharge at Rochester during these floods is estimated at about 40,000 second 
feet from a drainage area of 2.3G5 square miles. The discharge in the present flood can not yet be estimated. 
It is probable that the extensive damage and high stage readied in Rochester are due iu part to constriction of 
the channel, as the Genesee River flows through the heart of the city, and, in fact, passes underneath the main 
business street in the heart of the commercial district, where the river is covered over by buildings for a 
considerable distance. Just above this street is located the Erie Canal Aqueduct, the obstruction of which 
has hitherto been a prolific source of trouble in causing floods. 



SUPPLEMENTAL NOTE ON FREQUENCY OF RECURRENCE OF HUDSON 

RIVER FLOODS. 



By Robkrt E. Horton, Consulting Hydraulic Engineer. 



The question of the frequency with which a flood of a given magnitude may be expected 
to occur is probably of fully as great importance as the physical data relative to the flood 
itself; yet this matter of flood frequency seems to have been generally overlooked. The enthu- 
siastic advocate of storage for flood control or the man whose property has recently suffered 
damage from a severe flood is not apt to stop to think that the particular flood under considera- 
tion may not occur again in 100 years. The man who has been injured knows that the flood has 
occurred and apparently can occur again, and he naturally wants a remedy applied at once, if 
possible. The question of average frequency of recurrence of great floods is, however, of much 
importance. The question arises: Is the expenditure of large sums of money for the construc- 
tion of storage reservoirs justifiable to control floods which will occur on an average not oftener 
than once in say 100 years? There are few instances in which any reliable deductions as to the 
probable frequency of recurrence of great floods can be made. Since floods are generally due to 
excessive rains, the problem may be attacked by studying the frequency of recurrence of great 
storms. This is a valuable aid to the study for the reason that records of rainfall are available 
which are much more complete and of very much longer duration than records of the flood stages 
of streams. There is great lack of consistency in the practice of engineers in regard to provision 
for floods in engineering structures. On a given stream we will find one structure with flood 
provision adequate to withstand floods of the greatest intensity which will occur on an average 
once in 100 years, other structures, which are liable to injury by flood as often as once in 
10. 20, or 30 years on an average. 

The late George W. Rafter suggested to the writer in 1896 the possibility of the applica- 
tion of the theory of probabilities to the analysis of flood records, rainfall records, and other 
hydrological data. The object in view w'as the development of a consistent method for the 
design of engineering structures dependent for their success on hydrological phenomena of 
more or less periodic occurrence. Following this suggestion, the writer has developed and 
applied methods of analysis of such data by the theory of probabilities, utilizing in some 
instances the ordinary Gaussian law and in some instances developing special probability curves, 
formulas, and diagrams similar to those here given for the Hudson River. 1 The above state- 
ment is made in order to give Mr. Rafter credit for the first suggestion of the method in its 
broadest sense and also because the method has been used by the writer for more than 10 years 

1 Not reproduced. 

109 



110 THE FLOODS OF 1913. 

iii connection with the design of structures aggregating several millions of dollars in cost, and 
the further application of this method seems to be justified by actual experience. 

It is certainly in the line of making the best possible use of the available information. 
This method has been applied to flood discharges of the Hudson River at Mechanicsville. Table 
1 shows the date in each year on which the maximum flood discharge has occurred at Mechanics- 
ville during the past 26 years. The quantity of discharge given is the greatest average for one 
day. The absolute maximum for less than one day's time was somewhat greater in each instance. 
In applying this method to the Hudson River the data have been arranged in the order of 
magnitude shown in column 2 of Table 2. Column 3 shows the number of observed floods equal 
to or greater than a given magnitude. Column 1 contains the reciprocals of the numbers in 
column 3 and shows the probable average interval of recurrence in years of floods equal to or 
greater than the corresponding quantities in column 2. The data from column 2 have been 
plotted as ordinates and those from column 3 as abscissas on the accompanying diagram. 1 Loga- 
rithmic cross-section paper was used for plotting, and the ordinates were adjusted by the use 
of an addition constant, so as to reduce the line of plotted points substantially to a straight 
line. A straight line drawn through the plotted points represents empirically the law of 
relation between magnitude and frequencj^ of recurrence of floods within the range of observa- 
tion. By the extension of this line or by deduction of the corresponding formula the probable 
frequency of recurrence of floods greater than any hitherto observed can be inferred. In making 
such inferences it must of necessity be assumed that the same laws govern the frequency of 
recurrence of greater floods. Two formulas for this special probabilit}' curve are given on the 
diagram. One formula shows the magnitude " Q " of floods which will probably be equaled or 
exceeded once in " I " years on an average ; the other shows the average interval of years " I " 
at which the flood of a given magnitude " Q " will probably be equaled or exceeded. 

I _/Q + 50,000\7.14 
V 80,000 / 

Q = 80,000 I - 14 -50,000 

For example: In order to determine the probable average interval of recurrence of a flood as 
great as that of March, 1913. we have 

j /l 13,500 + 50,000y 14 



Solving by logarithms, we get 



80,000 / 



1=165 vears. 



Here we have a flood magnitude which will apparently only be equaled or exceeded once 
in 1G5 years on an average, but which has actually been observed once in a period of 26 years. 
There are a number of ways in which data similar to that here given may be plotted to deduce 
a special probability curve. It may be plotted either as a direct or as an inverse probability 
curve and either on logarithmic, semilogarithmic, or plain paper. It is often desirable to plot 
the data by two or more methods in order to fix the best position of the curve. This was done 
in the present instance. 

If the occurrence of floods of different magnitudes was a matter of pure chance, then their 
relative frequenc)' of recurrence could be determined by the ordinary Gauissian law of error. 
In cases where the available records are of short duration the cleriviation of a special probability 
curve by the method which has been given is impracticable. In such cases the Gaussian law of 
error properly applied is a valuable aid in the interpretation of the data. As a rule, however, 
the writer has found that large floods occur with much greater frequency than the Gaussian 
law would indicate. This is accounted for in part by the fact that in the derivation of the 

1 Not reproduced. 



FREQUENCY OF RECURRENCE OF HUDSON RIVER FLOODS. 



Ill 



Gaussian law, phis and minus departures from the mean of the observed phenomena, are 
assumed to be equally probable. Obviously a flood can not very well be as much as 100 per 
cent smaller than the mean flood, while it may exceed the mean flood by '200 or 300 per cent 
or even more. It follows that in a given record there are usually more floods below tbe mean 
than there are above. The average maximum yearly discharge of the Hudson River at 
Mechanicsville for the past 26 years has been 41,870 second feet. In 17 out of the 26 
years the maximum discharge was less than this, and it was greater in only !> years. As a 
result of this condition, the Gaussian law indicates departures above the mean having con- 
siderably less than the observed or actual frequency. This error may be partly eliminated in 
applying the Gaussian law by considering only departures above the mean; but even so. (he 
Gaussian law indicates a probable flood interval of recurrence for a flood on Hudson River 
as great as that of 1913 of 667 years, as compared with an average interval of 105 years deduced 
from a special probability curve. 

Without going into detail, the writer would say that the analysis of many of the longest 
rainfall and flood records throughout the world points strongly to the conclusion, tentatively 
at least, that floods of very great intensity actually occur with considerably more frequency 
than the laws of occurrence of floods of ordinary intensity would indicate should be the case. 
To what this is due can not at present be stated. The suggestion is inevitable, however, that 
conditions come into play in the production of heavy rainfalls which are not ordinarily in 
operation. The ultimate causes of rainfall are much less fully understood to-day than was 
supposed to be the case 10 or 20 years ago. Recent investigations of molecular physics, con- 
densation on electrons and ions, electrification, surface tension,. and the possibility of supersatu- 
ration at high altitudes suggest a number of ways in which excessive rainfalls may be produced 
at rare intervals from causes which are not ordinarily in operation. Owing to this apparent 
tendency for abnormally great floods and rainfalls to occur with greater frequency than the laws 
of occurrence of lesser floods would indicate should be the case, the waiter is of the opinion that 
floods of this magnitude in the Hudson River may probably be expected to recur at average 
intervals less than 165 years. That a flood as great as this, exceeding the average flood for 
26 years by 170 per cent and exceeding the greatest flood hitherto recorded by 62 per cent, will 
be of rare occurrence seems obvious. 

Table 1. — Maximum discharge of Hudson River at Mechanicsville, JV. Y., for 26 years. 

[Drainage area 4,500 square miles. Maximum given is greatest average flow for 24 hours.] 



Date. 



Discharge 
(cubic feet 
per second) 



Apr. 7, 1888. 
May 1, 1889. 
May 27, 1890 
Apr. 20, 1891 
Apr. 25, 1892 
Apr. 15, 1893 
Mar. 24, 1894 
Apr. 10, 1895 
Apr. 19, 1896 
Dec. 16, 1897 
Mar. 14, 1898 
Apr. 26, 1899 
Apr. 23, 1900 
Apr. 24, 1901 
Mar. 3, 1902. 



31, 130 
22, 400 
24, 300 
33, 120 
30, 110 
25. 460 
25, 560 
49, 630 
67.000 
35, 700 
39, 231 
41,475 
43, 546 
54, 862 
41.362 



Cubic feet 

per second 

per square 

mile. 



6.93 
4.99 
5.40 
7.14 
6.69 
5.66 
5.68 
11.0 
14.9 
7.93 
8.72 
9.22 
9.68 
12.2 
9.19 



Mar. 25, 1903. 
Apr. 11, 1904. 
Apr. 1, 1905. . 
Apr. 6, 1906. . 
Apr. 1, 1907.. 
Apr. 28, 1908. 
Apr. 16, 1909. 
Apr. 3, 1910. . 
May 3, 1911 . . 
Apr. 8, 1912. . 
Mar. 28, 1913. 



Average . 
Spring, 1869.... 



Date. 



Discharge 
(cubic feet 
per second) 



56, 283 
36,305 
48, 877 
40,279 
36, 672 
34,335 
46, 299 
37,809 
26, 241 
47. 275 
113,510 



41,875 
70. 000 



Cubic feet 

per second 

per square 

mile. 



12.5 
8.07 

10.9 
8.95 
8.15 
7.63 

10.3 
8.41 
5.83 

10.5 

25.2 



15.6 



112 



THE FLOODS OF 1913. 



Table 2.— Floods of Hudson River at Meclianicsvillc N. Y., arranged in order of magnitude, ivith average 
interval of recurrence and departures from the mean for each flood. 



Number. 


Discharge 
(cubic feet 
per second). 


Average 
interval 
(years). 


Departure from the 
mean. 


Departures for floods 

exceeding 41,876 cubic 

feet-seconds. 




Excess. Deficiency. 


Excess. 


Deficiency. 


1 


2 


3 


4 5 


6 


7 


1 


22.400 


1.00 
1.04 




19. 474 
17.574 
16.414 
16.314 






2 


24.300 








• 
3 25. 460 1. 09 








4 25. 560 1. 13 










26,241 1.18 15.633 

30,110 1.24 11.764 






6 








31.130 1.30 


10. 744 
8,754 
7,539 
6,174 
5,569 
5,202 
4,065 
2,643 
1.595 






s 


33.120 1.37 
34. 335 1. 45 













Ill 


35. 700 


1.53 
1.62 






11 36 305 








12 . 


36, 672 


1.73 
1.86 
2.00 
2.17 








13 


37.809 
39.231 

40.279 








14 








15 








16 


41,362 2.36 
41.475 2.60 
43.546 2.89 
46.299 l 3.25 
47,275 1 3.71 
4S.877 ; 4.33 
49. 630 5. 20 
54. 862 6. 50 
56, 283 8. 67 
67.000 , 13.00 
113.510 26.00 




512 






17 




400 






IS 


1,672 
4, 425 




15.041 


19 






11. 2^S 


20 


5.401 
7.003 
7, 756 
12,988 
14.409 
25, 126 
71.636 






11.312 


21 







9,710 


2' 




8,957 


23 






3.725 


24 






2.304 


25.. . 




8,413 




26 




54. 923 




Mean 






41.876 













Flood record rainfall at Conservation Commission stations in New York State. 



Date. 


5 
| 

< 


> 

o 
o 

- 


<fl 


S> 

c 

o 


a 
3 


Z 


a 

s 

o 

1 


o 

ft 

C 


© 

5 


*© 
= 
■a 
© 


_© 
> 

S3 

E 

a: 


© 

C 
5 


O 

M 

•3 
O 


> 

© 

a 

C3 

s- 

c 


.a 

■r. 

:>. 

> 


i 

a 
•a 
3 
a 


"© 




1 


o 


3 


4 


5 


6 


.1 


8 


9 


10 


11 


12 


13 


14 


15 


16 


17 


18 


19 


1913. 






































Mar. 22 
























0.16 
.28 


0.25 


0.35 
1.41 


0.60 
1.94 








23 










0. 75 


0.06 








0.41 


0.70 


0.50 


0.50 


0.19 


0.38 





24 


0.25 


0. 15 


0.45 


.70 


.18 


0.77 


.18 


1.35 


.99 


1.03 


.97 


1.68 


1.03 


.S3 


.40 


.S5 


2.00 


0.41 


25 


1.2S 


1.10 


.67 


.61 


1.41 


.99 


2.00 


.70 


.58 


1.30 


.60 


.42 


.91 


1.17 


1.20 


.70 


1.05 


1.10 


26 


.61 


.36 


1.31 


.25 


.99 


. 58 


1.25 


1.70 


1.24 





1.18 


1.13 


2.20 


.42 


.23 


.90 


1.79 


.S2 


27 


1.17 
.21 



1.52 



.26 



1.30 



.24 
T. 


1.29 

.OS 


1.55 



.06 





.06 


.11 


T. 


.02 






.50 




1.00 


28 









20 






















30 


























22-2^ 


3.52 


3. 13 


2.69 


3.61 


2.88 


3.71 


4. 9S 


4.22 


3.51 


2.^9 


3. 30 


3. (17 


4.41 


4. IS 


4.47 


3.14 


5. 22 


3.33 



FLOODS IN THE CONNECTICUT VALLEY AND IN VERMONT, MARCH 

25-31, 1913. 



By W. W. Netfert, Local Forecaster, Hartford, Conn. 



The mild rainy weather of March 20-22 reduced to water the small amount of snow and 
ice remaining in the woods and mountains of the extreme upper watershed. Under usual sea- 
sonal conditions this small quantity of snow and ice would not have resulted in great floods; 
but the fact that considerable frost yet remained in the ground and the thoroughly saturated 
condition of the soil, made conditions favorable for a rapid run-off. The small streams soon 
filled to overflowing, breaking up the ice and causing gorges. These conditions being aug- 
mented two days later by heavy rains, a sharp rise in the larger streams quickly followed. The 
ice gorges and overflowing streams caused a moderate amount of damage in the small streams 
at Montpelier, Barre, Lancaster, Littleton, Lyndonville, and St. Johnsbury, Vt. Moreover, by 
reason of the fact that the initial stages of the larger streams were relatively high, a large 
volume of water was soon sweeping down the upper Connecticut with irrsistible force, 
inundating thousands of acres of land, and doing damage of such extent that it can not be 
accurately estimated in a monetary sense. The flood was due chiefly to heavy rain falling on 
thoroughly saturated soil, which was only partly free from frost, and consequently many of the 
washouts and landslides occurred on frost formations. The absence of the usual spring covering 
of snow with its considerable water content, surely minimized the losses which occurred.. How- 
ever, the flood over the upper valley was the greatest since 1869, while in the lower valley it 
brought the highest water since 1896. 

On the headwaters of the Connecticut, as at Wells River, Vt., the river was at a relatively 
high stage on the morning of the 25th. It rose 3 feet, viz, from 27 to 30 feet during that 
clay, and this rise, in connection with the rains which had fallen over the watershed, was the 
first intimation of a flood throughout the course of the river in Massachusetts and Connecticut. 

At White River Junction, 46 miles below Wells River, the Connecticut River at 8 a. m. 
March 25, stood at 14.6 feet, and at 4 p. m. of that date it had risen to 20.3 feet ; it reached a 
maximum height of 30 feet at 9 p. m. of the 27th, thus overtopping all previous high records 
back to f 869. While no precise measurements of the stages of the river at White River Junc- 
tion during the memorable floods of 1869 and 1862 are at hand, a reliable witness of both floods 
is authority for the statement that the 1913 flood exceeded both of the earlier floods. 

The breaking of a big log boom at Sharon, Vt. (on White River), by which 2,500,000 to 
3,000.000 feet of logs were set adrift, was the cause of the loss of the highway bridge at White 
River Junction, and the placing in jeopardy of the Boston & Maine Railway bridge across the 
White River. 

At Holyoke, Mass., the river was at a stage of 5.9 feet at 8 a. m. of the 26th, 9.2 at the cor- 
responding hour of the 27th, and crested at a stage of 12 feet at 7 p. m. of the 28th, continuing 
at that stage practically all of that night, The high water not only flooded many of the 
cellars in the lower portions of the city, but also caused the suspension of power plants and 
manufacturing concerns during the continuance of the flood. 

14284°— 13 8 113 



114 



THE FLOODS OF 1913. 



At Springfield, Mass., the highest stage readied was 20.8 feet, or nearly 2 feet below high 
water of 1854. At this stage cellars were flooded, streets and railways and low-lying truck 
farms were submerged, and in some cases persons were compelled to move to the upper stories of 
their houses in order to escape the water. The same experience was had at Thompson ville and 
Windsor Locks, points farther down river. 

At Hartford persons along the river front began moving their belongings to places of 
safety as soon as a flood stage was forecast; when 20 feet of water on the Hartford gage was 
assured an immense amount of movable property was transferred to higher levels; nevertheless, 
much damage to buildings and other immovable property resulted. 

Below is a comparative table of 1913 maximum stages as compared with earlier records. 

Comparative table of flood stages. 



Station. 



Wells River, Vt 

White River Junction, Vt 

White River June! ion (White River) 

Bellows Falls, Vt 

Holyoke, Mass 

Springfield, Mass 

Hartford, Conn 



Highest 
stage on 
record. 


Year. 


Feet. 




35.0 


1909 


26.3 


1895 


25.7 


1895 


19.4 


1895 


12.7 


1869 


22.2 


1854 


'29.8 


1854 



Highest 
stage 1913. 



Feet. 



36.0 
30.0 
29.4 
19.0 
12.0 
20.8 
26.3 



i The 1869 (lood made a mark of 26.7 feet. 

The loss to railroads in connection with this flood is estimated at $200,000. As there ai'e 
many things that can not be reduced to dollars and cents which are really of considerable 
damage to railroad and other interests, it is impossible to furnish exact data as to losses. In 
Hartford the loss ranges from $50 in some cases to $1,000 in others, and likewise in the various 
towns the losses range from $100 in one town to $2,000 in others. 

Note. — Loss of bridge at White River Junction amounts to $37,000; value of logs of Sharon boom actually 
lost, $10,000; loss to Hartford tobacco firms, $2,000; value of freight house at Middletown, $7,000: loss to 
steamboat company at Hartford, $2,500. 

The fatalities were : A man drowned at Highgate Springs, Vt. ; a boy drowned while pleasure 
boating at Hartford; and at East Putney, Vt., a railway employee was drowned when a freight 
train slid into the river. 



INDEX. 



A. 

Acknowledgments 9 

Alciatore, II. F., report of flood by 5] 

Alps, II. F., report of flood by 97 

Atchafalaya River, report of flood in 101 

15. 

Brand, Al., report of flood by 85 

Barron, W. E. , report of flood by 93 

C. 

Cade, W. R., report of flood by '. 75 

Connecticut River, report of flood in 1 13 

Crest stages in the Ohio in floods of 1884-1913 30 

( 'revasses 40, 90, 94, 96, 1 02 

Crevasse near — 

Beulah, Miss 40 

Briers, La 102 

Columbus, Ky 90 

Graves Bayou, Ark 91 

Lake St. John, La 102 

North Memphis, Tenn 91 

Poydras, Miss 102 

Skipworth, Miss 91 

West Hickman, Ky 91 

Wilson, Ark 91 

D. 

Dayton, Ohio, flood at 51 

Devereaux, W. C. , explanation by 25 

E. 

Emery, S. O, report of floods by 89 

Evansville district, Ohio River, report of flood in 85 

F. 

Flood in the Ohio : 

At Cairo 26 

Cincinnati to Louisville 26 

Compared with previous floods 30 

January and February, 1913 : 15 

Louisville to Cairo 26 

March and April, 1913 15 

Parkersburg to Cincinnati 25 

Flood stages 12 

Floods : 

Classification of 12 

Contributing causes of 11 

Frequency of — 

By periods 36 

In the Ohio River 1 . 33 

In Hudson River, frequency of recurrence of 1 09 

In Illinois, report on 77 

In Indiana rivers, report on 49, 7 1. 75 

In October, 1910 '..... 32 

115 



116 INDEX. 

Floods — Continued. 

I n Ohio— p age . 

Before systematic observations began 14 

During period 1870-1910 15 

Five feet or more above flood stage 13 

Of 1806 14 

Of 1832 14 

Of 1847 14 

In Ohio River — 

By groups 14 

During last 40 years 12 

In Ohio rivers, report on 45 

In the lower Mississippi River, March, 1913 39 

Duration of, as compared with 1912 " 40 

Midsummer 33 

Midwinter 32 

Probable maximum in the Ohio River at Pittsburgh 37 

G- 

Gauge readings: ' 

Hourly at Parkersburg. W. Va., Mar. 26-31, 1913 25 

In the Ohio watershed 27-28 

Great Miami River, hydrograph of 1913 flood 51 

H. 

Hamilton, Ohio, flood at 55 

Hayes, Montrose W. . report of flood by 77 

Hocking River, report of flood in 49 

Horton, Robert E.: 

Flood report by 105 

Report on frequency of recurrence of Hudson River floods 109 

Hudson River, flood in 105 

Hudson River floods '. . . 109 

Hydrographs for floods of 1913 105 

Hydrographs of Great Miami flood 51 

I. 

Illinois River, report on flood in 77 

Indiana: 

Floods in rivers of 49 

Flood in — 

White River '. 71 

Wabash River 75 

L. 

Land overflowed 40, 41 

Loss due to floods 41, 42 

M. 

Marietta, Ohio, flood at 29 

Memphis, Tenn., report of flood at 89 

Meteorological condition as affecting floods 11, 12 

Mahoning River, flood in ] 47 

Manmee River, flood in 47 



Xeifert , W. W. . report on flood in Vermont and Connecticut Rivers, by. 113 

Norquest, C. E. . report on flood in White River, by 77 

New Orleans district, report of flood in 101 

O. 
Ohio River: 

Cincinnati to Louisville 26 

Comparison of 1913 flood with previous floods 29-30 



INDEX. 117 

Ohio River — -Continued. 

Flood in — Page. 

January, 1913 15 

March-April, origin of 16 

Floods in, 1870-1913 11 

Louisville to Cairo 26 

Parkersburg to Cincinnati 25 

Physical features of the basin 11 

Pittsburgh to Parkersburg 24 

Precipitation: P. 

Charts of — 

Mar. 23-27, 1913 28 

Jan. 1-13, 1913 28 

Oct. 3-6, 1910. . 28 

Feb. 4-7, 1884 28 

Daily amounts of, in Ohio watershed — 

Mar. 23-27, 1913 20-23 

Feb . 4-14, 1884 31 

Excessive, at Cincinnati 16 

In Ohio- 
Charts of 45 

Mar. 23-27 45 

Progressive averages of 35 

Secular variation in 35 

S. 

Smith, Prof. J. Warren, report on floods by 45 

Snow, on ground in February, 1884 31 

T. 

Temperature, daily record of, in Ohio watershed, Feb. 1-15, 1884 30 

Todd, G. T., report on flood in Hudson River by 105 

U. 

XJniontown, Ky. , loss at 87 

V. 

Vicksburg, river district, floods in 93 

W. 

"Wabash River, Ind., flood in 75 

Walz, Prof. F. .1. . report on floods in Ohio by , 81 

Warnings - 42, 73, 83, 86, 92, 101, 105 

White River, flood in 71 

o 



LBJe'14 



mESAEX 0F CONGRESS 



029 708 376 4 






H 



■ i> 






^m 






■ 



■ 









■ 

■ 



.h 









■ 

i 



■ 



I 

■ 

JA 



i •<